Sadler Apps Unit 4

Australia • Brazil • Japan • Korea • Mexico • Singapore • Spain • United Kingdom • United States
Mathematics Applications Unit 4 © 2017 A J Sadler

1st revised Edition
A J Sadler Copyright Notice
This Work is copyright. No part of this Work may be reproduced, stored
in a retrieval system, or transmitted in any form or by any means
Publishing editor: Robert Yen without prior written permission of the Publisher. Except as permitted
Project editor: Alan Stewart under the Copyright Act 1968 , for example any fair dealing for the
Cover design: Chris Starr (MakeWork) purposes of private study, research, criticism or review, subject to
Text designers: Sarah Anderson, Nicole Melbourne, certain limitations. These limitations include: Restricting the copying to
Danielle Maccarone a maximum of one chapter or 10% of this book, whichever is greater;
Permissions researcher: Kaitlin Jordan providing an appropriate notice and warning with the copies of the
Answer checker: George Dimitriadis Work disseminated; taking all reasonable steps to limit access to these
Production controller: Erin Dowling copies to people authorised to receive these copies; ensuring you hold
Typeset by: Nikki M Group Pty Ltd the appropriate Licences issued by the Copyright Agency Limited
(“CAL”), supply a remuneration notice to CAL and pay any required
Any URLs contained in this publication were checked for fees. For details of CAL licences and remuneration notices please
currency during the production process. Note, however, that contact CAL at Level 15, 233 Castlereagh Street, Sydney NSW 2000,
the publisher cannot vouch for the ongoing currency of Tel: (02) 9394 7600, Fax: (02) 9394 7601
URLs. Email: [email protected]
Website: www.copyright.com.au
For product information and technology assistance,

in Australia call 1300 790 853;
in New Zealand call 0800 449 725
For permission to use material from this text or product, please email
[email protected]
National Library of Australia Cataloguing-in-Publication Data

Sadler, A. J., author.
Mathematics applications. Unit 4 / A J Sadler.
1st revised edition

9780170395069 (paperback)
For secondary school age.
Mathematics--Study and teaching (Secondary)--Australia.

Mathematics--Problems, exercises, etc.
Mathematics--Textbooks.
Cengage Learning Australia

Level 7, 80 Dorcas Street
South Melbourne, Victoria Australia 3205
Cengage Learning New Zealand

Unit 4B Rosedale Office Park
331 Rosedale Road, Albany, North Shore 0632, NZ
For learning solutions, visit cengage.com.au
Printed in China by China Translation & Printing Services.

1 2 3 4 5 6 7 21 20 19 18 17
PREFACE
This text targets Unit Four of the
West Australian course Mathematics
Applications, a course that is
organised into four units altogether,
the first two for year eleven and the
last two for year twelve.
This West Australian course,
Mathematics Applications, is based on the Australian
Curriculum Senior Secondary course General
Some chapters commence with, or contain,
Mathematics. At the time of writing there are only small
a ‘Situation’ or two for students to consider, either
differences between the Unit Four content of these
individually or as a group. In this way students are
two courses, the main difference being that specific
encouraged to think and discuss a situation, which
examples mentioned in the Australian Curriculum are
they are able to tackle using their existing knowledge,
moved to the glossary of the West Australian syllabus,
but which acts as a forerunner and stimulus for the
or removed. Hence this text is also suitable for anyone
ideas that follow. Students should be encouraged to
following Unit Four of the Australian Curriculum course,
discuss their solutions and answers to these situations
General Mathematics.
and perhaps to present their method of solution to
The book contains text, examples and exercises others. For this reason answers to these situations
containing many carefully graded questions. are generally not included in the book.
A student who studies the appropriate text and
At times in this series of books I have found it
relevant examples should make good progress
appropriate to go a little beyond the confines of the
with the exercise that follows.
syllabus for the unit involved. In this regard readers
The book commences with a section entitled will find that in this book I briefly revise simple interest
Preliminary work. This section briefly outlines before proceeding to compound interest and, on
work of particular relevance to this unit that students the basis that centred moving averages might be
should either already have some familiarity with assumed content under the general inclusion of moving
from the mathematics studied in earlier years, or for averages, I include mention of this idea. (However
which the brief outline included in the section may be if one considers a centred moving average as
sufficient to bring the understanding of the concept up a weighted average then the glossary for the course,
to the necessary level. at the time of writing this text, says that weighted
As students progress through the book they will moving averages are beyond the requirements for
encounter questions involving this Preliminary work in the unit.) I leave it up to the readers and teachers
the Miscellaneous exercises that feature at the to decide whether to cover these aspects or not.
end of each chapter. These miscellaneous exercises Alan Sadler
also include questions involving work from preceding
chapters to encourage the continual revision needed
throughout the unit.
ISBN 9780170395069 iii

C NTENTS
PRELIMINARY WORK vii
1 3
Number...................................... vii
Percentages................................. vii
TIME SERIES DATA 2 FINANCE I: SAVING
Straight line graphs .......................viii AND BORROWING 52
Time series data ............................ 5
Solving equations .......................... ix Revision of simple interest,
Making predictions ........................ 6
Simple and compound interest .......... x compound interest and
Miscellaneous exercise one ........... 16
Recursion ...................................... x recursive rules ............................. 53
Graphs or networks ....................... xi Effective annual interest rate........... 58
2 Initial deposit plus regular
Bivariate data ..............................xiii
investments ................................. 60
Regression ..................................xiv
MOVING AVERAGES Compounding isn’t only
Use of technology.........................xvi
AND SEASONAL EFFECTS 18 about investing............................ 64
The statistical investigation
Making underlying trends Loans with regular repayments ....... 65
process .................................... xviii
more apparent ............................ 20 What price house can these
Algorithms ................................. xviii
Moving averages ........................ 21 people afford?............................ 70
General smoothing of time Miscellaneous exercise three ......... 72
series data ................................. 26
Use of spreadsheets ..................... 27
Centred moving averages ............. 28
Quantifying the seasonal effect ...... 33
Deseasonalising, or seasonally
adjusting, the data....................... 39
Making predictions ...................... 40
Miscellaneous exercise two ........... 48
iv MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

4 6 8
FINANCE II: DRAWING MAXIMUM FLOW 108 ASSIGNMENT

DOWN THE INVESTMENT 76 Maximum flow PROBLEMS 140
Superannuation ........................... 77 – a systematic approach............. 110 Assignment problems.................. 142
Annuities .................................... 79 Maximum flow = minimum cut...... 115 Allocating for a maximum ........... 143
Indexing .................................... 80 Miscellaneous exercise six........... 117 Four machines, four tasks ............ 144
Perpetuities ................................. 81 An algorithm for solving
Suppose the frequency of payments allocation problems ................... 146
7
≠ the frequency of compounding .... 84 Suppose we want a maximum ..... 151
Miscellaneous exercise four ........... 89 Suppose the number of people
PROJECT NETWORKS 120
≠ the number of tasks ................. 157
Project networks ........................ 122
Miscellaneous exercise eight ....... 163
5 Constructing a project network ..... 129
Miscellaneous exercise seven ...... 136
ANSWERS .............................. 173
MINIMUM
SPANNING TREES 92 INDEX ..................................... 203
Spanning trees ............................ 93

Minimum spanning trees
– a systematic approach............... 94
Minimum spanning tree from
the distances table ....................... 99
Miscellaneous exercise five ......... 105
ISBN 9780170395069 Contents v

IMPORTANT
N TE
This series of texts has been written based on Acknowledgements
my interpretation of the appropriate Mathematics
As with all of my previous books I am again
Applications syllabus documents as they stand at the indebted to my wife, Rosemary, for her assistance,
time of writing. It is likely that as time progresses
encouragement and help at every stage.
some points of interpretation will become clarified
and perhaps even some changes could be made to the To my three beautiful daughters, Rosalyn,
original syllabus. I urge teachers of the Mathematics Jennifer and Donelle, thank you for the continued
Applications course, and students following the course, understanding you show when I am ‘still doing sums’
to check with the appropriate curriculum authority and for the love and belief you show.
to make themselves aware of the latest version of I thank my good friend and ex-colleague Theo
the syllabus current at the time they are studying Wieman for his much appreciated wise counsel.
the course.
To the delightfully supportive team at Cengage -
I thank you all.
Alan Sadler
vi MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

PRELIMINARY W RK
This book assumes that you are already familiar with a number of mathematical ideas from your mathematical
studies in earlier years.
This section outlines the ideas which are of particular relevance to Unit Four of the Mathematics Applications
course and for which some familiarity will be assumed, or for which the brief explanation given here may be
sufficient to bring your understanding of the concept up to the necessary level.
Read this ‘preliminary work’ section and if anything is not familiar to you, and you don’t understand the brief
mention or explanation given here, you may need to do some further reading to bring your understanding of
those concepts up to an appropriate level for this unit. (If you do understand the work but feel somewhat ‘rusty’
with regards to applying the ideas some of the chapters afford further opportunities for revision as do some of
the questions in the miscellaneous exercises at the end of chapters.)
• Chapters in this book will continue some of the topics from this Preliminary work by building on the
assumed familiarity with the work.
• The miscellaneous exercises that feature at the end of each chapter may include questions requiring an
understanding of the topics briefly explained here.
Number
The understanding and appropriate use of the rule of order, fractions, decimals, percentages, rounding,
truncation, square roots and cube roots, numbers expressed with positive integer powers, e.g. 23, 52, 25,
expressing numbers in standard form, e.g. 2.3 × 104 (= 23 000), 5.43 × 10 7 (= 0.000 000 543), also called scientific
-
notation, and familiarity with the symbols >, ≥ , < and ≤ is assumed.
Percentages
It is assumed that you are able to express one quantity as a percentage of another quantity and find a percentage
of a quantity. It is also assumed that you are able to increase or decrease an amount by a certain percentage by
appropriate multiplication. For example:
• To increase an amount by 25% we multiply by 1.25.
• To decrease an amount by 5% we multiply by 0.95.
Typical percentage questions:

I Express 28 out of 40 as a percentage.
28
First express as a fraction:
40
28
Then multiply by 100: × 100 = 70%
40
Thus 28 out of 40 is 70%.
ISBN 9780170395069 Preliminary work vii

II Find 28% of $40.
$40
Either: Divide by 100 to find 1%:
100
$40
Then multiply by 28 to find 28%: × 28
100
= $11.20
Or: Use the decimal equivalent of 28 %
28% of $40 = $40 × 0.28
= $11.20
The following examples of ‘typical percentage questions’ will only show the ‘decimal equivalent’ method.
III Increase $40 by 28%.

$40 increased by 28% = $40 × 1.28
= $51.20
IV Decrease $40 by 28%

$40 decreased by 28% = $40 × 0.72
= $28.80
IV 28% of an amount is $40. Find the amount.

28% of an amount is $40
$40
100% of the amount =
0.28
= $142.86 (nearest cent).
Straight line graphs

If points lie in a straight line their coordinates, ( x, y), obey a rule of the form
y = mx + c
and the relationship between x and y is said to be linear .
In this rule m is the gradient of the straight line and the point (0, c) is where the line cuts the vertical axis.
Hence the line y = 2x + 1 has a gradient of 2 and cuts the y-axis at (0, 1).
The line y x + 5 has gradient of - 1 and cuts the y -axis at (0, 5).
= -
In the equation y = mx + c the value of m, the gradient, tells us the amount by which y increases for each unit
increase in x.
viii MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

Solving equations
It is assumed that you are already familiar with the idea that solving an equation involves finding the value(s) the
unknown can take that make the equation true.
For example, x = 5.5 is the solution to the equation 15 - 2x = 4
because 15 - 2(5.5) = 4.
To solve an equation, e.g. 15 - 2x = 4, we could proceed mentally :
We know that fifteen take eleven equals four.
Thus 2x =11 and so x = 5.5.
We could use the solve facility of some calculators.
Or we could use a step by step approach to isolate x:
Given the equation: 15 - 2x = 4
Add 2x to both sides to make the x term positive: 15 = 4 + 2x
Subtract 4 from both sides to isolate 2x : 15 - 4 = 2x
\ 11 = 2x
Divide both sides by 2 to isolate x: 5.5 = x
\ x = 5.5
It is also anticipated that you are able to solve more involved linear equations, for example:
5x - 7 - 2x - 8 = 30 - 6x Solution: x = 5
x+3
- 1 = 5 Solution: x = 9
2
2(x + 3) - 3(2x + 1) = - 5 Solution: x = 2
and simultaneous linear equations, for example:

 3x + 2 y = 11 
 5x - 2y = 6 
 2x + 3 y = 12



x + 2y = 1 


3x + 2y = 26



x + 4 y = 11
Solution: x = 5, y = -2 x = 4, y = 7 x = 3, y = 2
In this unit we may also encounter equations involving exponents (powers), for example
x4 = 81 2x = 64 2x = 11
Solve these mentally if the numbers involved allow it, otherwise use the solve facility of some calculators.
x4 = 81 2x = 64 2x = 11
x = 3 x = 6
solve(2 x 11, x)=
{ x 3.459431619}
=
x = 3.46 (correct to 2 decimal places)
ISBN 9780170395069 Preliminary work ix

Simple and compound interest
Suppose we were to invest $500 for 3 years in an account paying interest at the rate of 10% per annum.
Considering a simple interest approach in which the same amount of interest is added each year:
Value after 3 years = $500 + 3 × 10% of $500
= $500 + 3 × $50
= $650
Alternatively, considering compound interest, in which the interest earned in one year itself earns interest in
the next year:
Value after 1 year = $500 + 10% of $500
= $500 × 1.1
= $550
Value after 2 years = $550 × 1.1 (i.e. $500 × 1.12)
= $605
Value after 3 years = $500 × 1.13
= $665.50
If we want the value after n years:
Value after n years = Initial value × 1.1n
Recursion
It is assumed that you are familiar with the idea of defining a sequence by a recursive rule or recurrence
relationship, i.e. a rule which tells us how the terms of the sequence recur. For example, the following rule
T1 = 5, Tn + 1 = Tn + 3 (or T1 = 5, Tn = Tn - 1 + 3)
tells us that the first term of the sequence is 5 and each term after that is obtained by adding 3 to the previous term
Hence the sequence is: 5, 8, 11, 14, 17, 20, 23, …
With adjacent terms having a common difference this is an example of an arithmetic sequence, also called an
arithmetic progression or AP.
Thus all arithmetic sequences are of the form:
a, a + d, a + 2d, a + 3d, a + 5d, a + 4d, a + 6d, …
In this general form we have a first term of ‘a’ and common difference ‘d’.
The nth term is then given by Tn = a + (n - 1)d.
Sequences which progress such that each term of the sequence is the previous term multiplied by some constant
number are said to be geometric sequences, geometric progressions, or GPs.
Thus all GPs are of the form:
a, ar, ar 2, ar 3, ar 4, ar5, ar6, …
In this general form we have a first term of ‘a’ and common ratio ‘r’.
The nth term is then given by Tn = a × r n - 1
.
GPs will have a recursive rule of the form T1 = a, Tn + 1 = r × Tn.
x MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

It is also assumed that you are familiar with recurrence rules of the form:
T1 = a, Tn + 1 = b × Tn + c.
This type of recursive rule will generate sequences for which each term, after the first, is obtained from the
previous term by multiplying by some constant value, b, and then adding a constant value, c.
For example T1 = 4000, Tn + 1 = 1.01 × Tn - 100.
Thus T1 = 4000, T2 = 1.01 × T1 - 100
= 1.01 × 4000 - 100
= 3940
T3 = 1.01 × T2 - 100
= 1.01 × 3940 - 100
= 3879.4
T4 = 1.01 × T3 - 100
= 1.01 × 3879.4 - 100
= 3818.194 etc.
Recurrence relations of this form are particularly relevant to this unit.
Graphs or networks
In mathematics, a network or graph is a set of points, called vertices, connected by a set of lines, called edges.
1 2
• The junctions and endpoints are called vertices (singular: vertex). 4 vertices
4 3
• The connections are called edges. 5 edges 2 4 5

3
Graphs may be directed or undirected.
B
A C
A C
A directed graph An undirected graph

(Also referred to as a digraph)
The directed edges are sometimes
referred to as arcs.
ISBN 9780170395069 Preliminary work xi

Some graphs show ‘weights’ on each edge, perhaps distance, cost, time etc.
Such graphs are referred to as weighted graphs. For example:
B
120 km
75 km 63 km D
84 km 42 km
87 km
A C F
108 km
30 km
185 km
E
Whilst it is anticipated that from your study of Unit Three of Mathematics Applications you are familiar with
many of the terms associated with graph theory (e.g. multiple edges, adjacent vertices, simple graphs, connected
graphs, planar graphs, subgraphs, bridges, walks, paths, cycles, etc.), in this unit we will be mainly focusing on
analysing weighted graphs and in one such aspect an understanding of the word tree in the context of graph
theory is important, and in another aspect bipartite graphs are significant. These two terms in particular are
further explained below:
Any connected simple graph that contains no cycles is called a tree:
The following graphs are trees:
The following are not trees:
Contains a cycle. Not connected. Contains a cycle.

Therefore not a tree. Therefore not a tree. Therefore not a tree.
xii MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

Consider the following graph:
A B C
D E
Notice that if we consider the five vertices A, B, C, D and E as being in two sets:
{A, B, C} and {D, E}
then every one of the five edges joins a vertex from one set to a vertex of the other set.
AD, AE, BD, BE, CE.
A graph in which the vertices can be split into two groups such that every edge joins a vertex from one group
to a vertex of the other group, i.e. no edges join vertices from the same group, is called a bipartite graph. Thus,
in a bipartite graph, every pair of adjacent vertices involves one vertex from one group and the other vertex from
the other group.
For example, the bipartite graph below shows which of the three areas, midwifery (Mid), accident and emergency
(A & E), and the intensive care unit (ICU), three nurses are qualified to work in.
Rosemary Chris Taylor
Mid A&E ICU
Bivariate data
If a situation involves two data sets, for example the heights and weights of a number of people, we have
bivariate data. We might then be interested in investigating whether there is an association between the
data sets, for example, is there some association between a child’s age and their foot size? Do older children
generally have bigger feet? Does a child’s age in any way explain their foot size? In this case, age would
be the explanatory variable, also called the independent or predictor variable. Foot size would be the
response variable, also called the dependent or predicted variable.
nasutniP reeB/moc.kcotsrettuhS
ISBN 9780170395069 Preliminary work xiii

y
Regression 300
The scattergraph on the right shows the consumption
)noitalupop fo 000 001 rep raey reP(

of a particular food commodity on the horizontal x -axis,
esaesid traeh ot eud shtaeD

and heart disease death rate on the vertical y -axis, for
twenty countries. 200
The location of the points suggests that as one of the

measures increases then the other does too.
This apparent correlation between the two variables 100
could be modelled by drawing a line of best fit or

trend line, as shown on the next graph.
10 20 30 40 50 60 70
x
Foodstuff consumption
(Kilograms per person per year)
With one variable generally increasing as the other y

increases this line of best fit has a positive gradient. 300
We say that a positive linear correlation seems )noitalupop fo 000 001 rep raey reP(
to exist.
esaesid traeh ot eud shtaeD
The line of best fit allows predicted values to 200

be suggested.
For example we could suggest that a country with
an intake of the foodstuff of 45 kilograms per person 100
per year would have a heart disease death rate of
approximately 230 persons per year per 100 000
of population.
x
10 20 30 40 50 60 70
Foodstuff consumption
(Kilograms per person per year)
Rather than having to draw this ‘line of best fit’

by eye we can obtain its equation and predict Linear Reg
values using the linear regression techniques a = 3.09976491
of some calculators and spreadsheets. b = 90.894086
The display on the right gives the equation of the r = 0.82280944
2
line of best fit, or least squares regression line, as r = 0.67701539
y = ax b
y = 3.10x + 90.9
+
xiv MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

Given that the data for the previous scattergraph is as shown below make sure that you can obtain this
equation from your calculator.
x 57 20 13 38 51 48 58 35 29 8 28 21 10 40 41 46 53 64 66 47
y 299 124 154 156 152 295 264 208 250 84 158 175 120 220 251 202 248 290 296 268
We can then calculate (or determine automatically from our calculator) a predicted value for ‘y’ (in this case
deaths per year per 100 000 of population) for an ‘x’ value (in this case foodstuff consumption) of 45.
y = 3.0998(45) + 90.8941 i.e. approximately 230.
To emphasise the predicted nature of y this is sometimes written as ŷ, pronounced ‘y hat’. Thus for x = 45,
ŷ ≈ 230.
The ‘r ’ given in the previous calculator display is called Pearson’s correlation coefficient. It will lie in the
interval -1 ≤ r ≤ +1. This correlation coefficient is a measure of the strength and nature of the linear association,
or correlation, between the variables. It informs us how closely the relationship between two sets of data
approximates to a linear relationship.
A correlation coefficient of - 1 indicates a perfect linear relationship with a negative gradient (as one variable
increases the other decreases) and a correlation coefficient of + 1 indicates a perfect linear relationship with
a positive gradient (as one variable increases the other increases).
(The display also shows r2, the coefficient of determination. This gives the proportion of the variation
that can be explained by the linear relationship. I.e. in the previous situation the r2 value of 0.677 means that
approximately 68% of the variation in death rates due to heart disease can be explained by, or accounted for,
by the variation in the foodstuff consumption.)
If there is an association between the two variables then as one changes we should expect the other to change,
but that is not to say that the changes in one variable cause the changes in the other. There could be other
factors at work.
If we use the fact that a strong correlation seems to exist, to predict a ‘y-value’ for some ‘x-value’, we can be more
confident of this prediction if interpolation is involved, i.e. if the x-value used is within the range of x-values
covered by the given data.
If instead we are extrapolating, i.e. using an x-value outside the range covered by the x-values in the given data,
we would be less confident with any prediction made, our confidence decreasing the further the x -value is from
the range of x -values in the given data.
ISBN 9780170395069 Preliminary work xv

Use of technology
You are encouraged to use
• calculators to solve equations when necessary, to display the terms of a recursively defined sequence when
appropriate, and to display graphs,
• spreadsheets on a computer (or calculator),
• the internet for research.

 x y 1200
+ =



7 x 10y 9150 x, y
+ =
{ x 950, y 250}
= =
A B C D E F G
1 Physics score Maths score
2 6 19 Physics and Mathematics scores
3 9 12
50
4 11 27 45 y = 1.9829x + 4.8291
scitamehtaM ni erocS
5 12 36 40
35
6 14 42 30
7 14 34 25
20
8 16 32
15
9 16 27 10
10 17 44 5
0
11 19 41 0 5 10 15 20 25
12 Score in Physics
13
sneviN yegreS/moc.kcotsrettuhS
xvi MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

Calculators and computers can be particularly useful if we wish to display the terms of sequences. Consider for
example the growth in the value of a house that is initially valued at $500 000 and is subject to an annual increase
in value of 6.4%.
Initial value = $500 000 ← T1
Value after 1 year = $500 000 × 1.064 ← T2
Value after 2 years = $500 000 × 1.064
2
← T3
3
← T4 etc.
These values form a geometric sequence with Tn + 1 = Tn × 1.064
and T1 = 500 000
We could display the terms of our sequence on a calculator or spreadsheet and view the progressive year by year
values, as shown below.
an+1 = an ·1.064
A B C D
a1 = 500000
1 Initial value $500,000.00
bn+1 =
2 Percentage increase 6.40
b1 = 0
3 Value at end of year 1 $532,000.00
cn+1 =
4 2 $566,048.00
c1 = 0
5 3 $602,275.07
6 4 $640,820.68
n an
7 5 $681,833.20
8 6 $725,470.52 9 8.2E +5
9 7 $771,900.64 10 8.7E +5
10 8 $821,302.28 11 9.3E +5
11 12 9.9E +5
9 $873,865.63
13 1.1E +6
12 10 $929,793.03
13 11 $989,299.78
1052614.96443211
14 12 $1,052,614.96
These tables allow us to see, for example, that the value of the house will reach one million dollars shortly after
the end of the eleventh year, i.e. early in the 12th year.
However note carefully that in this situation, with the recursive definition
Tn + 1 = T n × 1.064 and T1 = 500 000
T1 is the value after zero years. Hence we must remember that if we use the ability of a calculator to generate
the terms of the sequence, according to the recursive rule given above, then the value after n years will be given
by Tn 1. I.e. in the calculator display above, n = 13 gives the value at the end of 12 years.
+
ISBN 9780170395069 Preliminary work xvii

One way to avoid this possible source of confusion would be to use
the ability of some calculators to accept a sequence defined using an+1 = an ·1.064
T0 as the 1st term. a1 = 500000
I.e. define the sequence as: bn+1 =

b1 = 0
Tn + 1 = Tn × 1.064 and T0 = 500 000,
cn+1 =
as shown on the right. c1 = 0
Under such a definition Tn would indeed be the value after n years.
n an
Alternatively we could use
Tn + 1 = Tn × 1.064 and T1 = 500 000 × 1.064 8 8.2E +5
9 8.7E +5
and again Tn would be the value after n years. 10 9.3E +5
11 9.9E +5
12 1.1E +6
1052614.96443211
The statistical investigation process

Given some real world problem, for example, estimating the future health needs for a particular region with
regards to number of hospital beds required, the statistical investigation process follows the following steps:
(1) Clarify the problem and formulate one or more questions that can be answered with data.
(2) Design and implement a plan to collect and obtain appropriate data.
(3) Select and apply appropriate graphical or numerical techniques to analyse the data.
(4) Interpret the results of this analysis and relate the interpretation to the original question; communicate
findings in a systematic and concise manner.
(As stated in the Australian Curriculum Glossary for General Mathematics and the West Australian Glossary for Mathematics Applications.)
Algorithms
Remember, an algorithm is a defined set of steps that that can be applied and systematically followed to
solve a particular type of problem. In Unit Three of Mathematics Applications we encountered an algorithm
for determining the shortest path between two vertices in a network.
xviii MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

1.
Time series data
• Time series data
• Making predictions
• Miscellaneous exercise one
Situation One
In the year 2015 Mrs Maggie Smithson sells
a valuable painting at auction for $1 243 000.
Mrs Smithson’s accountant informs her that for
taxation purposes he needs to know what the
value of the painting was when it first came into
her possession.
draW kcirtaP/otohP kcotS ymalA

Mrs Smithson says that it was left to her in
2001 upon the death of the previous owner,
Mrs Smithson’s aunt, but she did not know the
value of the painting at that time. What she did
have though was a number of valuations, carried
out at various times for insurance purposes.
The accountant listed these insurance valuations, together with the price achieved at auction as follows:
Year of valuation 1948 1968 1988 2008 2015

Valuation $85 000 $280 000 $345 000 $810 000 $1 243 000
Suggest what you consider to be a reasonable valuation for the painting for the year that it first
became the property of Mrs Smithson and explain how you arrived at your estimate and why you
think it reasonable.
Situation Two
A group is asked to write a report outlining the
future health needs for a particular region. One
of the things the group needs is an estimate of
the number of hospital beds the region is likely
to need in 5, 10 and 15 years time.
The number of hospital beds available in the
anoab/moc.kcotSi
region over the last twenty years, recorded
every 2 years, is known and is shown in the
following table:
Time 20 years ago 18 years ago 16 years ago 14 years ago 12 years ago 10 years ago
Number of
3270 3350 3510 3520 3760 3900
hospital beds
Time 8 years ago 6 years ago 4 years ago 2 years ago Now
Number of
4100 4420 4790 5140 5600
hospital beds
Assuming the above numbers did just about meet the needs of the community at the time, use the
figures to make predictions for the number of hospital beds this region is likely to require in 5, 10 and
15 years’ time and explain how you arrived at your predictions.
ISBN 9780170395069 1. Time series data 3

The situations of the previous page each involve measurements of something where the time the
measurement is made is also recorded. In each situation the given information is an example of
time series data.
Each of the situations involved the statistical investigation process:
(1) Clarify the problem and formulate appropriate questions.
In Situation One we needed to determine the value of the painting when it came into
Mrs Smithson’s possession. Not knowing this information we needed to ask what was known
about its value at other times.
In Situation Two we needed to determine the number of hospital beds required for this region in
5, 10 and 15 years’ time. To determine this we needed to ask for historical data about the number
of beds required in the region.
(2) Plan the collection and use of appropriate data.

In Situation One the plan was to use previous insurance valuations we were given.
In Situation Two the plan was to use previous data that was available regarding bed numbers.
(3) Apply appropriate techniques to analyse the data.

This was your task.
Did you consider graphing the information given in each of the situations?
Graphing time series data can allow underlying trends to be seen and predictions to be made.
(The trend is the general direction of the time series, e.g. increasing, decreasing, etc.)
(4) Interpret the results in order to answer the original question and communicate your findings.
This also was your task.
If you did graph the data of the two situations you may have come to the conclusion that attempting to
fit a straight line to the data might not be a suitable choice of ‘model’. Perhaps you tried another model
or, with each set of points perhaps suggesting a suitable curve that could be drawn ‘freehand’ maybe
you drew such a curve and read off the required values, or maybe you simply joined adjacent points
with straight lines.
One of the situations required us to extrapolate beyond the known values so predictions obtained
in this way would have to be viewed with caution. However, though extrapolation can be unreliable,
it is sometimes the best we can do. With the situation that did not require extrapolation, but instead
required us to determine a value between known points (interpolation), our prediction should be more
reliable. However, even with interpolation, out-of-the-ordinary events and short-term fluctuations can
sometimes give values that are not in line with normal trend. For example we might expect monthly
visitor numbers to a city to peak higher than a general trend would suggest if the city were to host an
Olympic Games for a month. We might expect daily traffic flow along a particular major road to drop
below the figure a general trend would suggest if there were major roadworks taking place along the
road and diversions were in place, etc.
4 MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

Time series data
Average price of 1 bedroom unit in
The diagram on the right shows a particular area, 1970 to 2015
a line graph, i.e. a graph in which $300 000
each point is joined to the next by
a straight line. $250 000
This particular line graph involves $200 000

observations that are ordered
according to time, in this case in five-
ecirP
$150 000
year intervals from 1970 to 2015.
The graph involves time series data. $100 000
Note that when graphing time

$50 000
series data, time is plotted on the
horizontal axis.
1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
The joining of the points allows the
Year
underlying trend to be seen.
The graph shows an increasing trend – the prices increase with the passing of time.
Drawing straight lines between adjacent points can allow underlying patterns to become more evident.
In the graph shown below there is clearly an increasing trend but can you also see the five point
repeating pattern?
Time Time
We will consider repeating patterns like this in more detail in the next chapter.
It can sometimes be the case that time series data is collected at intervals that are not all of the same
length (as was the case in Situation One). Consider the following situation for example:
An environmental group, concerned with the levels of a particular pollutant in a river, collected and
analysed samples from the river over a period of some years. The results are shown in the table below.
Month Jan Jul Jan Jul Jan Jul Jan Apr Jul Oct Jan Apr Jul
and year 2010 2010 2011 2011 2012 2012 2013 2013 2013 2013 2014 2014 2014
Pollutant
level 2.04 2.09 2.24 2.51 2.44 2.76 2.84 2.92 3.06 3.04 3.14 3.21 3.37
(units/m3)
Notice that in 2010, 2011 and 2012 levels were recorded half-yearly, but in 2013 levels were recorded
quarterly. The graph of this data is shown on the next page.

4
) m/stinu( level noitulloP

3
3
2010 2011 2012 2013 2014

Ja Ap Jul Oc Ja Ap Jul Oc Ja Ap Jul Oc Ja Ap Jul Oc Ja Ap Jul Oc
Time
• Notice that the horizontal axis uses a consistent time scale throughout. The fact that from January
2013 onwards pollutant levels are given 4 times per year, rather than the 2 per year prior to this,
changes the number of points plotted per year not the scale on the axis.
Making predictions
If we want to analyse the data mathematically we might need to number the dates involved, rather than
to attempt putting Jan 2010, Jul 2010 etc. into our calculator.
Particular care needs to be taken if the frequency of data collection changes as in this pollution levels
example. Noticing that from 2013 onwards pollutant levels are given 4 times per year we could number
the data points as follows:
Data point
1 3 5 7 9 11 13 14 15 16 17 18 19
(n)
Month Jan Jul Jan Jul Jan Jul Jan Apr Jul Oct Jan Apr Jul
and year 2010 2010 2011 2011 2012 2012 2013 2013 2013 2013 2014 2014 2014
Pollution (P)
2.04 2.09 2.24 2.51 2.44 2.76 2.84 2.92 3.06 3.04 3.14 3.21 3.37
(units/m3)
The graph would then be as shown below:
3
) P( level noitulloP
1 3 5 7 9 11 13 14 15 16 17 18 19
Time (n)

With the data following a reasonably straight line we could then use our knowledge of linear regression
to determine the least squares regression line:
P = 0.0735n + 1.91
(and a correlation coefficient of 0.99).
Asked to predict pollution levels for January 2018 we note that this would be for when n has the value 33.
For this value of n the regression line gives a predicted value of 4.3 for P.
Thus the predicted pollution level for January 2018 is 4.3 units/m3.
However, even with our ‘eye-balling’ of the points as being reasonably linear, and the correlation
coefficient of 0.99 indicating that using a linear model is appropriate, January 2018 is well beyond
dates during which pollution readings were collected so the reliability of the prediction is extremely
questionable. If the trend continues it may be a reasonable prediction but from July 2014 to Jan 2018
many things could occur to alter the trend.
Of course in some cases linear regression may not be suitable - as in the two situations at the beginning
of this chapter. In such cases other models available on some calculators could be used, and you are
encouraged to explore such possibilities, or simply ‘eye-ball’ the data and continue the trend as seems
appropriate. Whilst in this text we will concentrate on using a least squares linear regression model it is
clearly important to be able to recognise situations for which linear regression is not suitable.
Consider again the earlier
$500 000
situation involving the price of
a 1 bedroom unit in a particular
$450 000
area. In this case linear regression
would be inappropriate because $400 000
a straight line model would not
fit the data points at all well. $350 000
If asked to use the graph to
predict what the average price $300 000
was likely to have been in 2008
ecirP
we could draw the line up from $250 000
2008 until it meets our graph and

$200 000
then draw across, as shown on the
right, to obtain an estimate, in
$150 000
this case approximately $156 000.
Similarly, if asked to estimate the $100 000
likely average price for the unit in
the year 2020 we could continue $50000
the trend of the graph, as best we
can, and read off a value, in this
1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020
case approximately $480 000.
Year
However, as we know, this latter
situation again involves extrapolation
so any prediction could not be
considered to be particularly reliable.

EXAMPLE 1
The table below shows the percentage of the Australian population aged 80 or over, for the years
1995 to 2007.
Percentage of the Australian population aged 80 and over
Year 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Percentage 2.6 2.6 2.7 2.8 2.8 2.9 3.1 3.2 3.3 3.3 3.4 3.5 3.6
[Source of data: Australian Bureau of Statistics.]
Display this information as a line graph and comment on any trends shown.
Use your graph to suggest what percentage of the Australian population will be aged 80 or over
in the year 2020 and in the year 2050, each time discussing the likelihood of your prediction
being reliable.
Solution
Percentage of Australian
The graph is shown to the right. 4.0 population aged 80 or over
3.5
noitalupop fo egatnecreP
3.0
2.5
2.0
1.5
1.0
0.5
’95 ’96 ’97 ’98 ’99 ’00 ’01 ’02 ’03 ’04 ’05 ’06 ’07
Year
The percentage of the Australian population aged 80 or over shows an increasing trend.
For the period the data describes, i.e. 1995 to 2007, the percentage increases fairly steadily at a rate
of approximately 0.1% of the Australian population per annum. (Figures show increase of 1%
of population in 12 years.)
Continuing the trend forward to 2020, just over another 12 years, would suggest approximately
4.6% or 4.7% of the population would be 80 or over by then. With the data showing a reasonably
linear trend we could use linear regression to obtain a 2020 figure of approximately 4.7%.
This prediction involves extrapolation so the result should not be considered particularly reliable
but considering the steadiness of the increase from 1995 to 2007 the estimate may prove to be
quite good.
The reasonably steady increase of about 1% every 12 years shown in the graph suggest
approximately 7.2% of the population would be 80 or over by 2050. (Linear regression gives
7.4%.) However this requires us to extrapolate so far beyond the known points the ‘prediction’
is perhaps more an ‘educated guess’ than a reliable prediction.
→ Try to obtain more up-to-date percentages and see how they compare. ←

Note: • Time series data can often be seasonal in nature. For example, in the pollution levels in
a river situation encountered earlier, the higher temperatures in summer may alter the levels
of some pollutants. Thus in addition to showing some overall trend this sort of data may
also show variation dependent upon the season when the data is collected. We will consider
the seasonal nature of some time series data in the next chapter.
• The vertical axis should start from zero if possible. Sometimes this
520
may be inconvenient or impractical and in such cases a break in the 510
axis should be clearly shown (see the diagram on the right). 500
• Remember that time is plotted on the horizontal axis and that you
should use the same time scale along the entire axis.
• If you have access to a computer with spreadsheet capability explore its ability, and the
ability of your calculator, to draw line graphs and use them to draw some of the graphs
requested in the next exercise.
• Remember that predicting into the future can be hazardous. Unanticipated events
can occur. Consider for example the graph below, which shows the Dow Jones Index
(a measure of the share values of a number of large companies in the USA) for the
beginning of March 2003 to the beginning of March 2008.
12 000
10 000
8 000
6 000
4 000
2 000
March March March March March March

2003 2004 2005 2006 2007 2008
These figures might lead us to expect that by the beginning of March 2009 the index would
be perhaps as high as 12 500. In fact it had fallen to just a little over 7000.
Events causing results that seem out of line with a suggested trend may not always be
unexpected. The graph on the next page shows the total expenditure, in billions of pounds,
of the people visiting London, from elsewhere in the UK, for the years 2008 to 2012.

Billion £
1.4
1.3
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
2006 2007 2008 2009 2010 2011 2012
[Source of data: London tourism report 2012/13. London & Partners.]
Does the somewhat higher than trend figure for 2012 indicate increasingly higher figures
for 2013, 2014 etc. are likely, or should it be expected with 2012 being the year of the
London Olympics?
Exercise 1A
1 Ageing population
The table below shows some real and some predicted figures for the number of persons aged over
65, per 100 persons of working age, in Australia for 1971, 1981, 1991, 2001, 2011, 2021 and 2031.
Number of persons over 65 per 100 persons of working age

Year 1971 1981 1991 2001 2011 2021 2031
Number 13 15 17 18 20 26 31
[Source of data: HBF.]
a We would of course expect the

number of people over 65 to increase
as the population of Australia
increases. Can the increase in the
numbers shown in the bottom row
of the table (i.e. 13, 15, 17, 18, …)
be attributed to this natural increase
in the population or not? Explain
your answer.
segamissenisubyeknom/moc.kcotSi
b Display the data as a line graph.

c Describe any trends shown.

2 Population of Western Australia
The following table gives the population (to the nearest 1000) of Western Australia in various years.
Year 1915 1925 1935 1945 1955

Population 317 000 378 000 450 000 490 000 669 000
Year 1965 1975 1985 1995 2005

Population 838 000 1 167 000 1 437 000 1 749 000 2 037 000
[Based on Australian Bureau of Statistics data.]
a Display this information as a line graph.

b Comment on any trends shown.
c By ‘eye-balling’ your graph, predict what the population was in the
year 1950 and comment on the likely reliability of your prediction.
Western
d By ‘eye-balling’ your graph, predict what the population will be in the Australia
year 2025 and comment on the likely reliability of your prediction.
e Find out the latest population figure for Western Australia and see
how it compares to that predicted by the above figures.
3 Visitors to Australia
The table below shows the number of overseas visitors arriving in Australia for a stay of less than
12 months (called short-term visitors ) for various years from 1992 to 2013.
Year 1992 1994 1997 2001 2003 2005

Visitors 2 603 100 3 361 600 4 318 000 4 855 800 4 745 800 5 463 000
Year 2007 2009 2010 2011 2012 2013

Visitors 5 588 800 5 490 200 5 790 100 5 770 800 6 032 200 6 380 500
a Display this information as a line graph

and comment on any trends shown.
b By ‘eye-balling’ your graph, predict the
number of short-term visitors to Australia
in 1999 and in 2020, in each case
commenting on the likely reliability Australia
of your prediction.

4 Age at first marriage
In more recent times is the age at which people get married for the
first time more or less than it used to be in ‘the old days’?
Use the data shown below to formulate a response to the above question, including in your
response a graphical display of the data, and commenting on, and describing, any trends shown.
Median age at first marriage in Australia

Year 1966 1970 1974 1980 1982 1984 1985 1988 1990 1991 1992
Men 23.8 23.4 23.3 24.2 24.6 25.1 25.4 26.1 26.5 26.7 26.9
Women 21.2 21.1 20.9 21.9 22.4 22.9 23.2 24.0 24.3 24.5 24.7
Year 1993 1996 1997 2000 2001 2002 2008 2009 2010 2011 2012
Men 27.0 27.6 27.8 28.5 28.7 29.0 29.6 29.6 29.6 29.7 29.8
Women 24.8 25.7 25.9 26.7 26.9 27.1 27.7 27.7 27.9 28.0 28.1
5 Number of marriages
Let us suppose that in a particular country, the number of marriages occurring in various years
from 1910 to 2010 were as shown in the following table.
Number of marriages occurring in the country in various years from 1910–2010

Year (Y) 1910 1925 1930 1938 1947 1962 1975 1986 1997 2004 2010
Number (N) 25 124 34 728 37 894 38 247 46 221 49 241 59 258 66 241 68 256 71 567 76 278
a View the graph of this time series data on a calculator or computer and confirm that the
points are suitable for linear modelling.
b With N representing the number of marriages in a year, and Y representing the year, obtain
the least squares regression line, N = mY + b , for this data, giving m to the nearest integer and
b to the nearest 1000.
6 Expanding business
The rental property manager of a real estate business researched data regarding the number of
properties managed by the business at various times from 2010 to 2015. The table below shows
the data she collected.
Mar Mar Mar Mar Mar Jun Sept Dec Mar Jun
Month and year
2010 2011 2012 2013 2014 2014 2014 2014 2015 2015
No. of properties
302 321 408 467 490 492 508 502 526 540
managed
a Commencing with March 2010 as t = 1, June 2010 as t = 2 etc. determine the equation of
the least squares linear regression line for predicting the number of properties managed for
a given value of t.
b Assuming this trend continues predict the number of properties managed by the group in
December 2017.

7 Caesarean sections
Let us suppose that in a particular region the number of deliveries for all births, over a sixteen-year
period, and the number of those deliveries that were performed by caesarean section, were as given
in the following table:
Year 1 2 3 4 5 6 7 8
Total births 15 750 16 170 15 971 16 227 16 451 16 332 16 815 16 213
Caesarean 2835 3017 2925 2973 3068 3612 3821 3659
Year 9 10 11 12 13 14 15 16
Total births 16 629 16 873 17 105 16 974 17 043 16 142 17 123 17 260
Caesarean 4083 4431 4862 4904 5082 4758 5161 5178
In year 1 the rate of caesarean deliveries was (16, 30)

2835 out of 15 750, i.e. 18%. 30
In year 16 the rate of caesarean deliveries was
) P( egatnecreP
5178 out of 17 260, i.e. 30%. 20
(1, 18)
If this increase in the caesarean rate over this
time were perfectly linear the caesarean rates 10
would fit the graph shown on the right.
a Determine the equation of this straight line, in
the form P = mx + c, with x and P as indicated 5
Year ( x)
10 15
on the graph, and m and c constants.

b Plot a graph of the real percentage figures over the 16 years to show that the points do not
form such a straight line and describe any trends the real percentage figures show over time.
8 Transport authorities in a particular country researched vehicle registrations over a number of

years to investigate the average age of the vehicles registered for road use. The results were:
Average age of vehicles
Year 1990 1992 1994 1996 1998 2000 2002
Average age (years) 7.6 7.8 8.4 9.1 9.6 10.2 10.5
Year 2004 2006 2008 2010 2012 2014

Average age (years) 10.6 10.7 10.7 10.6 10.4 9.9
a Display this data as a time series graph.

b Is the data suitable for linear regression? Justify your answer.
c Use your graph to predict the average age of vehicles in this country for the years 1999 and
2020, commenting in each case about the likely reliability of your prediction.
9 Confirm that the following paired values are not suited to linear regression.
t 1 2 3 4 5 6 7 8 9 10
Y 5 20 44 96 149 217 310 410 525 661
Explore the ability of graphic calculators to determine non-linear regression models to obtain
a model for the above data of the form Y = a × t b.

THE CHANGING SEASONS
Quite a number of middle-aged or elderly people have been heard to comment that they feel the
seasons have been changing over the years. For example, some feel that January tended to be hotter
in the past. Are they correct or is it that they are only remembering the more memorable summers?
The table below shows information about January temperatures recorded at a particular location
in Perth over a period of approximately 70 years. Does the information given in the table suggest
that, during the period of time the statistics cover, summer temperatures in Perth were showing
some changing pattern?
Write a brief report justifying your opinion and include suitable graphs based on the information
in the table below.
Temperature statistics for the month of January 1946–2014

Year 1946 1948 1950 1952 1954
Mean daily max temp (°C) 28.2 30.9 32.7 31.3 31.0
Mean daily min temp (°C) 13.3 15.3 16.8 14.8 17.6
Year 1956 1958 1960 1962 1964

Year 1966 1968 1970 1972 1974

Year 1976 1978 1980 1982 1984

Year 1986 1988 1990 1992 1994

Year 1996 1998 2000 2002 2004

Year 2006 2008 2010 2012 2014

[Source of data: Bureau of Meteorology.]

RESEARCH AND ANALYSIS
Research and obtain appropriate time series data to answer one of the following questions, or
negotiate with your teacher an alternative question of your choice that will similarly require an
analysis of time series data.
Your answer should include the data you obtained, suitably
presented, an explanation of the analysis of that data and
a reasoned conclusion that addresses the question.
kinreB reteP/moc.kcotsrettuhS
• When is the world record for running 100 metres
likely to fall to 9 seconds?
• When is the population of the world likely to reach
10 billion?
• What is the population of China likely to be in the
year 2050?
• What trends does Australia’s unemployment rate show over time?
• How has the share price of a particular company of your choice varied with time and what
might this suggest for the future price?
• How have house prices, or rental costs, varied over time in your area and what does this suggest
for future price, or rental, movements?
inamanelle/moc.kcotSi kriK nhoJ/moc.kcotSi oicavoneR/moc.kcotsrettuhS rednalojSnahoJ/moc.kcotSi

• How has the price of farming commodities varied with time
and what does this suggest for future price movements?
• How has the volume of air traffic changed with time

and what does this suggest for future air traffic volume?
• How have national or worldwide annual sales of new cars

changed with time and what does this suggest for future
numbers of sales?
• How have the estimated numbers of some endangered species of animal

varied with time and what does this suggest for the future of the species?

Miscellaneous exercise one
This miscellaneous exercise may include questions involving the work of this chapter and the
ideas mentioned in the Preliminary work section at the beginning of the book.
1 Increase $2050 by 13%.
2 Decrease $12 600 by 74%.

y
F B D A
3 Determine the equation of each of the 4
C
straight lines A to F shown in the graph 3
on the right.
2 E
−5 −4 −3 −2 −1 1 2 3 4 5 x
−1
−2
−3
4 Find the first five terms of each of the following recursively defined sequences.
a T1 = 25, Tn 1 = Tn + 2
+
b T 1 = 32, Tn 1 = Tn - 1.5
+
c T1 = 64, Tn 1 = 1.5 × Tn
+
d T 1 = 4000, Tn = Tn 1 ÷ 2
-
e T1 = 5, Tn = 2Tn 1 + 7
-
f T 1 = 16, Tn 1 = 1.5Tn - 4
+
5 Population changes
The table on the right shows the population of Perth, Population of WA and Perth
and of Western Australia as a whole, in each of the
Year WA Perth
census years from 1911 to 1991.
1911 282 114 116 181
Make a table of your own showing Perth’s population
1921 332 732 170 213
as a percentage of the WA population for each of these
census years. 1933 438 852 230 340
a Plot these percentages as a line graph, with years 1947 502 480 302 968
on the horizontal axis, and comment on the trends 1954 639 771 395 049
your graph shows. 1961 736 629 475 398
b Use your graph to suggest Perth’s population as 1966 836 673 558 821
a percentage of the WA population for i 1940 and 1971 1 030 469 703 199
ii 2020, in each case commenting on the likely
1976 1 178 340 832 760
reliability of your estimates.
1981 1 300 600 922 017
Try to find more recent data for the populations
1986 1 459 011 1 050 120
of Perth and of Western Australia, say for the census
years 2011 and 2016, and comment on how they compare 1991 1 636 067 1 188 762
with the data given here. [Source of data: Australian Bureau of Statistics]

2.
Moving averages
and seasonal
effects
• Making underlying trends more apparent
• Moving averages
• General smoothing of time series data
• Use of spreadsheets
• Centred moving averages
• Quantifying the seasonal effect
• Deseasonalising, or seasonally adjusting, the data
• Making predictions
• Miscellaneous exercise two
Situation
During 2014 the owners of a particular tourist attraction decided to use the data regarding the
number of visitors the centre has had in recent years to identify trends and to indicate likely visitor
figures for the future. The owners have access to the four-monthly visitor totals for 2011, 2012 and
2013 and also for the first four months of 2014. These figures are shown in the table below.
Year Period of time Visitors (thousands)

2011 During 1st 4 months 45
During 2nd 4 months 26
During 3rd 4 months 34
• Whilst these visitor figures seem to jump around

somewhat, is the underlying trend one of
increasing numbers of visitors,
decreasing numbers of visitors,
or a reasonably constant number of visitors?
segamInedruD/moc.kcotSi
• If the next figure, i.e. for the 2nd period of 4 months
of 2014, turns out to be 58, this is clearly a drop from
the 69 recorded for the 1st four months of 2014
but is it cause for concern?
The situation above asked you to do two things:

1. Attempt to identify an underlying trend in data that may initially appear rather erratic.
2. Think about the significance of a value beyond those initially given.
In this chapter we will consider

• making underlying trends more apparent,
and • making predictions,
for the sort of fluctuating data given above.
ISBN 9780170395069 2. Moving averages and seasonal effects 19

Making underlying trends more apparent
The graph below shows the data from the situation on the previous page.
Visitors (1000s)
70
60
50
40
30
20
Periods of
10
4 months
1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd 1st
2011 2012 2013
The graph shows the seasonal nature of the data with three ‘seasons’ making up one cycle.
The peaks occur at the 1st, 4th, 7th and 10th points plotted and the troughs at the 2nd, 5th and
8th points plotted. From this pattern we would certainly expect the next value, i.e. for the 2nd
4-month period of 2014, to be a trough. A figure of 58 for this 4-month period would therefore not
automatically be a cause for concern unless the drop were more than was expected.
(Note: Seasonal variation has a constant time period. A repeating pattern without a regular frequency
of occurrence would be called cyclical rather than seasonal.)
The graph indicates the general upward trend in the data as shown by the orange line below.
Visitors (1000s)
70
60
50
40
30
20
Periods of
10
4 months
2011 2012 2013
The placement of this line is just a ‘guesstimate’ based on ‘eye-balling’ the graph and guessing an
approximate line of best fit.
We could obtain a line of best fit using the linear regression techniques that you have encountered
previously but, with the original data being quite widely spread, and with an obvious seasonal nature,
it is better to first smooth the graph by attempting to remove the seasonal variation.
We could smooth the data by taking an average attendance for each of the three years but this will
reduce 10 data points down to just 3. Instead, because the given data has a three-point pattern to it
we create a 3-point moving average.

Moving averages WS
A 3-point moving average is found by first averaging the number of visitors for the
Smoothing time series
1st, 2nd and 3rd data points, data using moving
averages
then averaging the 2nd, 3rd and 4th data points,
then averaging the 3rd, 4th and 5th data points, etc.
In this way we move down the data points and each average will be formed using a 1st 4 months figure,
a 2nd 4 months figure, and a 3rd 4 months figure.
The first three 3-point moving averages are shown below (rounded to 1 decimal place when
appropriate). Note carefully the placement of the moving average (MA) figures.
Yr Months MA Yr Months MA Yr Months MA

11 1st 4 45 11 1st 4 45 11 1st 4 45
2nd 4 26 → 35 2nd 4 26 2nd 4 26
3rd 4 34 3rd 4 34 → 37.3 3rd 4 34
12 1st 4 52 12 1st 4 52 12 1st 4 52 → 41
2nd 4 37 2nd 4 37 2nd 4 37
3rd 4 42 3rd 4 42 3rd 4 42
The complete table is shown on the right.

Adding these values to the graph (see the orange dashed
Year Months Visitors 3-pt MA
line below) we see that this moving average technique has
2011 1st 4 45 –
smoothed the data considerably.
2nd 4 26 35
Visitors (1000s) 3rd 4 34 37.3
70 2012 1st 4 52 41
60 2nd 4 37 43.7
50 3rd 4 42 47.7
40 2013 1st 4 64 49
30 2nd 4 41 51.7
3-point moving average
20 3rd 4 50 53.3
10 Periods of
4 months 2014 1st 4 69 –
2011 2012 2013
Note The three-point moving average achieved this smoothing because the
original data had a three-point pattern to it. Had the original data had
The moving average has
a four-point pattern, as we might expect if we were collecting data
smoothed the seasonal
every quarter year, we would use a four-point moving average as
component in the data,
leaving the trend component
shown in example 1 on the next page. If data was collected each
more apparent. month we might expect a 12-point pattern. If data was collected
every day we might expect a 7-point pattern.

EXAMPLE 1
Year Period of time Number of cattle
To monitor the planned eradication of 2011 1st quarter 60
a particular disease in cattle all vets were 2nd quarter 67
required to notify the government of any 3rd quarter 47
4th quarter 38
cattle diagnosed as having the disease.
2012 1st quarter 52
The data collected was totalled every
2nd quarter 53
three months (or quarter of a year) and
3rd quarter 43
is shown on the right.
4th quarter 28
a Calculate the 4-point moving averages. 2013 1st quarter 42
b On a single graph plot both the original 2nd quarter 46
data and the 4-point moving averages. 3rd quarter 34
4th quarter 23
c Is the underlying trend decreasing, steady
2014 1st quarter 28
or increasing?
2nd quarter 32
Solution
a The 4-point moving averages are shown b The initial data and the 4-point moving
in the table below. Again note carefully averages are shown graphed below.
the placement of these figures.
Year Time No. of cattle 4-pt MA Number of cattle
2011 1st quarter 60 70
2nd quarter 67
60
→ 53
3rd quarter 47
51 50
4th quarter 38
47.5
40
2012 1st quarter 52
46.5
2nd quarter 53 30
44
3rd quarter 43
20
41.5
4th quarter 28 4-point moving average
39.75 10
2013 1st quarter 42
37.5 Year and quarter
2nd quarter 46 11 11 11 11 12 12 12 12 13 13 13 13 14 14
36.25 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2
3rd quarter 34
32.75 c The number of cattle diagnosed with the
4th quarter 23 disease is showing a decreasing trend.
29.25
2014 1st quarter 28
2nd quarter 32

Exercise 2A
For each of numbers 1 to 6 suggest the most appropriate moving average to use to smooth the given
set of data. (i.e. 3-point, 4-point, ? point.)
1 2
110
100 250
90
80 200
70
60 150
50
40 100
30
20 50
10
1 2 3 4 5 6 7 8 9 10 2 4 6 8 10 12 14
3 4
100 1000
90
80 800
70
60 600
50
40 400
30
20 200
10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10
5 6
400
70
60 300
50
40 200
30
20 100
10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5 6 7 8 9 10
Note for the remaining questions of this exercise:

Some calculators and computer programs, when given the original data, can be programmed to list
moving averages directly. However, even if your have access to such capability, do the remaining
questions of this exercise without using this facility, to ensure that you fully understand the process.

7 A drama group put on a particular play three times a week for five weeks.
The capacity of the theatre they use for the production is 500 and the numbers attending over the
five-week ‘season’ are given below right.
a Plot this data as a graph with Performance Week Day Attendance
the performance number on the number
horizontal axis and attendance
1 Thur 210
on the vertical axis.
2 1 Fri 320
b Add to your graph a plot of the data
3 Sat 370
after it has been smoothed using
a three-point moving average. 4 Thur 270
c Do the attendance figures indicate 5 2 Fri 440

that a sixth week of performances 6 Sat 450
would have been a good idea? 7 Thur 420
8 3 Fri 500
9 Sat 500
10 Thur 450
21namtah/moc.kcotSi
11 4 Fri 500
12 Sat 500
13 Thur 350
14 5 Fri 410
15 Sat 430
8 To monitor the planned eradication of Year Period Number of pigs

a particular disease in pigs all vets were required of time
to notify the government of any pig diagnosed 1 1st quarter 29
as having the disease. The data collected was
2nd quarter 31
totalled every three months (or quarter of
a year) and is shown on the right. 3rd quarter 21
a Calculate the 4-point moving averages. 4th quarter 25
b On a single graph plot both the original 2 1st quarter 28

data and the 4-point moving averages. 2nd quarter 24
c Is the underlying trend decreasing, steady 3rd quarter 21

or increasing? 4th quarter 23
3 1st quarter 26
2nd quarter 24
3rd quarter 19
4th quarter 20
4 1st quarter 25
2nd quarter 19
grubnosllit/moc.kcotSi

9 The table shown relates to the power bills received by a household every two months for a period
of three years.
a Determine values for the spaces
Year 2 month Bill ($) 6-pt
marked A, B, C, D and E. period MA
b On a single graph plot both the bill 1 2012 1st 2 months 238
amounts for the three years and
–
the six-point moving averages.
2 2nd 2 months 251
c Is the underlying trend decreasing,
–
steady or increasing?
3 3rd 2 months 283
258
4 4th 2 months 306
261
5 5th 2 months 274
A
6 6th 2 months 196
B
7 2013 1st 2 months 256
C
8 2nd 2 months 263
D
9 3rd 2 months 325
284
10 4th 2 months 342
287
11 5th 2 months 280
288
12 6th 2 months 238
700noelemaK/moc.kcotSi
287
13 2014 1st 2 months 274
289
14 2nd 2 months 269
290
15 3rd 2 months 319
293
16 4th 2 months 354
–
17 5th 2 months 286
–
18 6th 2 months E

General smoothing of time series data
Even if there is no obvious seasonal nature to time series data we can still use a moving average to
smooth out any irregular fluctuations and unsystematic peaks and troughs. For example consider
the following table showing the total wool production in Australia, in tonnes, every five years from
1915 to 2010.
Year 1915 1920 1925 1930 1935

Total production (tonnes) 291 300 300 800 330 700 425 600 460 500
Year 1940 1945 1950 1955 1960

Year 1965 1970 1975 1980 1985

Year 1990 1995 2000 2005 2010

Total production (tonnes) 1 049 800 679 400 641 500 475 200 352 700
(Based on Australian Bureau of Statistics data.)
Plotting this data below shows no particular systematic, calendar related movements but plotting
the moving average data (shown dashed) still has a smoothing effect and can make an underlying
trend clearer.
Production (tonnes)
1 100000
1 000000
900000
800000
700000
600000
500000
400000
300000
5 point moving average
200000
100000
Year
1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Use of spreadsheets
Spreadsheets can be particularly suitable for calculating moving averages and displaying the results
graphically. The information from the previous situation involving wool production in Australia can
be put into a spreadsheet, the moving averages can be calculated and the data displayed graphically,
as shown below.
A B C D E F G H I J K
1 Total Wool Production
2 Year Tonnes 5-pt MA
3 1915 291300 1200000
4 1920 300800
5 1925 330700 361780
1000000
6 1930 425600 405800
7 1935 460500 437640
8 1940 511400 474860 800000
9 1945 460000 505840
10 1950 516800 565840
11 1955 580500 625140 600000
12 1960 760500 715560

13 1965 807900 758040
400000
14 1970 912100 773580
15 1975 729200 771920
16 1980 658200 820300 200000
17 1985 752200 773760
18 1990 1049800 756220
19 1995 679400 719620 0
01
51
02
52
03
53
04
54
05
55
06
56
07
57
08
58
09
59
00
50
01
20 2000 641500 639720
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
02
02
02
21 2005 475200
22 2010 352700
23
Try to create your own spreadsheets to answer some of the questions in this chapter.
Note: You may find that a spreadsheet offers the facility of adding a trend line to a graph, and one
of the trend line options may be a moving average line for which you can set the period,
i.e. a 3-point, 4-point, … N-point moving average. However it may be the case that the moving
average points are plotted at the end of the period not at the centre, as we tend to show them.
Plotting values at the end of the values that are used to form the moving averages produces
the same line as shown dashed above, but moved to the right. In such a line the same values
are displayed ‘later’. The line is said to be showing the trailing moving averages. In this text we
will not use this line and will instead always plot our moving averages at the centre of the time
values used to form it, as shown above.
However, that is not to say that attaching a moving average to the end of the period used to
form it cannot be useful in some circumstances. Suppose for example that you were an athlete
monitoring your times in training. Today you might well form a three-point average involving
today’s time and the times of the previous two days. In your training diary you could well record
today’s figure as being the average of today’s time and that of the previous two days. I.e. you
would be recording the three-point moving average at the end of each three-day period used to
calculate it. But, as stated above, when plotting moving averages in this text in order to smooth
data we will plot at the centre of the time interval used to calculate it.

Centred moving averages Year Quarter No. of cattle 4-pt MA
Consider again the calculation of four-point 2011 1st 60

moving averages encountered previously
involving data collected with regard to a disease 2nd 67
in cattle, as shown on the right. Note the moving → 53
average figure does not align with a particular
3rd 47
quarter but instead falls between quarters.
This ‘misalignment’ of original data and moving 4th 38
average figures will occur in an N-point moving
average when N is even. For some uses of moving averages it is more convenient if the two quantities
do align. Hence it is quite common to give an N-point moving average centred when N is even. This
centring simply involves taking the average of two values, as the following table shows for the cattle
disease data.
t Year Quarter No. of cattle 4-pt MA Centred 4-pt MA
1 2011 1st 60
2 2nd 67
→ 53
3 3rd 47 → 52
51
4 4th 38 49.25
47.5
5 2012 1st 52 47
46.5
6 2nd 53 45.25
44
7 3rd 43 42.75
41.5
8 4th 28 40.625
39.75
9 2013 1st 42 38.625
37.5
10 2nd 46 36.875
36.25
11 3rd 34 34.5
32.75
12 4th 23 31
29.25
13 2014 1st 28
14 2nd 32

Note: • We can go directly from the initial data to the centred four-point moving average if we take
the five values that contribute to the centred average and sum ‘half the first, half the last,
and the middle three’ and then divide by four. Applying this idea to the previous example,
the first five values give
30 (i.e. half of 60) + 67 + 47 + 38 + 26 (i.e. half of 52)
= 52, as required.
4
The next set of five values give:
33.5 + 47 + 38 + 52 + 26.5
= 49.25, as required.
4
• Use a, b , c, d and e to prove that the above method gives the same result as is obtained by
first obtaining the uncentred 4-point averages.
• What would be the rule for centring six-point averages in this way?
Exercise 2B
1 The table below shows the number of cans of soft drink a lunch bar sells in each three months
of a three-year period.
a Why would determining 4-point moving averages be suitable for this data?
b Determine the values A, B, C, D, E, F and G.
t Year Quarter Cans sold 4-pt MA Centred 4-pt MA

1 One 1st 8312 –
–
2 2nd 5783 –
7006
3 3rd 5727 6983
6960
4 4th 8202 A
6920
5 Two 1st 8128 6906
B
6 2nd 5623 6840
C
7 3rd 5615 D
6738
8 4th 7786 6724
6710
9 Three 1st 7928 6698
6686
10 2nd E G
F
11 3rd 5519 –
–
12 4th 7594 –

2 The table shows the number of client interviews an accountant conducts in each 2-month period
over three years.
a Determine the values A, B, C, D E, F, G, H, I and J.
t Year 2-month period Units 6-pt MA Centred 6-pt MA

1 1 1st 142
2 2nd 235
3 3rd 295
175
4 4th 141 A
178
5 5th 80 C
B
6 6th 157 D
184
7 2 1st 160 183.5
E
8 2nd 247 F
185
9 3rd 319 185.5
186
10 4th 135 185
184
11 5th 92 181.5
179
12 6th 163 181
183
13 3 1st 148 185.5
188
14 2nd 217 I
H
15 3rd 343 J
193
16 4th 165
17 5th 104
18 6th G
b Reproduce the centred 6-point moving average figures using a spreadsheet showing only two
columns, the raw data ‘units’ column (with your value for G included) and the centred moving
average column. (See the note at the top of the previous page.)

3 Determine the moving averages for the information given below, which shows the tonnage of
a particular crop produced in each season, over two years. Take MAs to mean Moving Averages
and CMAs to mean Centred Moving Averages.
Year 1 Year 2
Summer Autumn Winter Spring Summer Autumn Winter Spring
1228 364 640 1220 1436 276 752 1132
4-point MAs
4-point CMAs
4 The monthly average price per kilogram of a particular type of fruit was monitored over a period
of four years. The data obtained is shown below.
Year 1
January February March April May June
$2.90 $2.70 $1.90 $1.80 $1.75 $1.90
July August September October November December
$1.90 $1.95 $2.00 $2.15 $2.60 $3.00
Year 2
$3.30 $2.40 $1.90 $1.95 $2.00 $1.95
$2.00 $2.10 $2.25 $2.35 $2.60 $2.80
Year 3
$3.20 $3.60 $2.60 $2.20 $2.05 $2.00
$2.00 $2.05 $2.10 $2.20 $2.60 $3.10
Year 4
$3.20 $2.65 $2.05 $1.95 $2.05 $2.20
$2.40 $2.50 $2.60 $2.90 $3.40 $4.30
a Why does it make sense to consider the 12-point moving averages for this data?
b Either manually, or with the assistance of a graphic calculator, spreadsheet or computer program
display the original data, and the 12-point centred moving averages, on a single graph.
c Is the underlying trend decreasing, steady or increasing?

5 If the population of a region increases we might expect the number of criminal incidents reported
by the police for that region to similarly increase.
For crimes reported by New South Wales police and classified as ‘Robbery with a firearm’ the numbers
reported annually from 1990 to 2013 were as follows (rounded to the nearest multiple of 5):
Number of incidents classified as ‘Robbery with a firearm’

recorded by NSW police, 1990 to 2013
Year Number Year Number Year Number

1990 980 1998 1120 2006 605
1991 1265 1999 890 2007 595
1992 1130 2000 805 2008 420
1993 875 2001 1105 2009 520
1994 520 2002 890 2010 445
1995 785 2003 835 2011 400
1996 1015 2004 710 2012 380
1997 1400 2005 520 2013 325
(© State of New South Wales through the Department of Justice and reproduced
with the approval of the NSW Bureau of Crime Statistics and Research.)
a Which would smooth the data more, a three-point moving average or a five-point
moving average?
b Display on a single graph the original figures, and whichever of the three- or five-point
moving averages you felt would smooth the data more.
c Comment on any trends shown.
Moving averages and finance

Moving averages are sometimes used by people monitoring the movement of share prices. A 30-day
moving average of share prices will be less influenced by a sudden change in the share price than
say a 5-day moving average would be. By considering and comparing different moving averages
a stock market analyst can attempt to identify trend and predict future movement in share prices.
In some cases the analyst might ‘weight’ the
moving average so that the more recent share
prices have more influence on the average than
a price from a few weeks ago. Whilst this idea
of a weighted moving average is beyond the
scope of this course, do a bit of research on the
internet to see what some financial sites say
hsilabula/moc.kcotSi
about the use of moving averages.

Quantifying the seasonal effect
Let us consider again the situation from the beginning of this chapter concerning visitor numbers to
a particular tourist attraction. These figures are shown again below.
Year Period of time Visitors (thousands)

Notice that there seems to be a ‘1st 4 months effect’ that leads to a higher number of visitors attending
the tourist attraction in the first four months than in the other two periods of four months in each year.
Consider the figures for 2011:

Total number of visitors in 2011 = 45 000 + 26 000 + 34 000
= 105 000
Average number of visitors per 4-month period = 105 000 ÷ 3
= 35 000
The 1st 4 month number, 45 000, as a percentage of this average figure, is 128.6% (1 decimal place).
The 2nd 4 month number, 26 000, as a percentage of this average figure, is 74.3% (1 decimal place).
The 3rd 4 month number, 34 000, as a percentage of this average figure, is 97.1% (1 decimal place).
Hence we could say that for 2011 the 1st 4 months have a seasonal effect that lifts the attendance
by 29% of the average, the 2nd 4 months have a seasonal effect that drops the attendance by 26%
of the average (74% is 26% below 100%) and the 3rd 4 months have a seasonal effect that drops
the attendance by 3% of the average.
However, with the figures for 2012 and 2013 available it would be better to carry out the above process
for all three years and then average the results. This is indeed how we will determine a measure of the
effect each season has – called the seasonal effect, seasonal component or seasonal index.
The three seasonal indices for the above data are calculated on the next page.

Year Period of time Visitors 4-month mean for Visitor numbers
(1000s) the year (1000s) as percentage of
4 month mean
2011 During 1st 4 months 45 35 128.57%
During 2nd 4 months 26 74.29%
During 3rd 4 months 34 97.14%
2012 During 1st 4 months 52 43.667 119.08%
2013 During 1st 4 months 64 51.667 123.87%
Calculation of seasonal effect, seasonal component or seasonal index:
1st 4 months 2nd 4 months 3rd 4 months

2011 128.57% 74.29% 97.14%
2012 119.08% 84.73% 96.18%
2013 123.87% 79.35% 96.77%
Mean (nearest %) 124% 79% 97%
The seasonal index for the first 4 months of a year is 124% (or 1.24), the seasonal index for the second
4 months of a year is 79% (or 0.79) and the seasonal index for the third 4 months of a year is 97% (or
0.97). This means that visitor numbers for the first four months of a year tend to be 24% above the
average 4 month numbers, for the second four months they tend to be 21% below the average 4 month
numbers and for the third 4 months of a year they tend to be 3% below the average 4 month numbers.
Note: • This particular method of determining the seasonal indices is called the average
percentage method. There are a number of other methods that are used to quantify
seasonal effects in time series data but in this unit we will use this average percentage method.
• The average of the seasonal indices should be 100%, i.e. with three seasons as shown above
the total should be 300%. This is the case for the above indices, 124 + 79 + 97 = 300.
If small errors due to rounding make an average not equal to 100, suitable adjustment
could be made.
• On occasions, though rarely in this text, the average percentage method uses the median
rather than the mean of the percentages. From the above table this would give indices of
123.87%, 79.35% and 96.77% (i.e. in this case it would make little difference, but that would
not always be the case). In this text, when asked for the seasonal indices to be calculated using
the average percentage method, the reader should assume the mean is to be used.

Read through the following which shows the calculation of seasonal indices for data involving six
seasons and make sure you understand the calculations involved.
t Year 2-month period Units 2-month mean Percentage of

for the year 2-month mean
1 1 1st 284 352 80.68%
2 2nd 463 131.53%
3 3rd 595 169.03%
4 4th 279 79.26%
5 5th 168 47.73%
6 6th 323 91.76%
7 2 1st 318 370 85.95%
8 2nd 493 133.24%
9 3rd 635 171.62%
10 4th 268 72.43%
11 5th 188 50.81%
12 6th 318 85.95%
13 3 1st 302 381 79.27%
14 2nd 428 112.34%
15 3rd 678 177.95%
16 4th 308 80.84%
17 5th 211 55.38%
18 6th 359 94.23%
1st 2nd 3rd 4th 5th 6th

2 months 2 months 2 months 2 months 2 months 2 months
1st year 80.68% 131.53% 169.03% 79.26% 47.73% 91.76%
2nd year 85.95% 133.24% 171.62% 72.43% 50.81% 85.95%
3rd year 79.27% 112.34% 177.95% 80.84% 55.38% 94.23%
Mean 82.0% 125.7% 172.9% 77.5% 51.3% 90.6%
The seasonal indices for the 1st 2 months, 2nd 2 months etc. are
1st 2 months: 82.0% (or 0.820), 2nd 2 months: 125.7% (or 1.257),
3rd 2 months: 172.9% (or 1.729), 4th 2 months: 77.5% (or 0.775),
5th 2 months: 51.3% (or 0.513), 6th 2 months: 90.6% (or 0.906).
Notice that 82.0 + 125.7 + 172.9 + 77.5 + 51.3 + 90.6 = 600, as we would want for data involving six
seasons as then the average index is 100. However, were we to round each index to an integer we would
have indices of 82, 126, 173, 78, 51 and 91 which give a total of 601, the ‘extra 1’ being introduced due
to the rounding carried out. To avoid this situation we could choose to round the 77.5 down to 77, to
give indices of 82, 126, 173, 77, 51 and 91. However in real applications we would probably be going on
to use these seasonal indices for further calculations, so we would be able to use more accurate values
stored in the cell of a spreadsheet, and would leave all rounding until the final calculation is completed.

Exercise 2C
1 Three of the seasonal indices for data with a quarterly seasonal pattern are as follows:
Seasonal index for the first quarter: 113%
Seasonal index for the second quarter: 105%
Seasonal index for the third quarter: 84%
Determine the seasonal index for the fourth quarter.
2 Six of the seasonal indices for data with a daily seasonal pattern are as follows:
Seasonal index for Monday: 73%
Seasonal index for Tuesday: 78%
Seasonal index for Thursday: 89%
Seasonal index for Friday: 102%
Seasonal index for Saturday: 143%
Seasonal index for Sunday: 121%
Determine the seasonal index for Wednesday.
3 A company expects the annual sales of one of their products next year to be 156 000 and, from
past years’ experience, anticipates no underlying increasing or decreasing trend, and no irregular
fluctuations in sales through the year.
a If there is no monthly seasonal effect on sales of this product roughly how many of the
product should the company expect to sell each month?
b In fact the sales are seasonal in nature with the seasonal indices for each month being
as follows:
January 88% February 82% March 76%

April 74% May 83% June 98%
July 104% August 105% September 108%
October 111% November 126% December 145%
Predict the sales numbers for this product for each month of next year.
4 A company finds that in the first quarter of the year it sells 17 360 units of a particular product.
From the quarterly sales of this product in previous years the company has calculated the seasonal
indices for sales of this product per quarter as being as follows:
Seasonal index for the first quarter: 112%
Seasonal index for the second quarter: 104%
Seasonal index for the third quarter: 98%
Seasonal index for the fourth quarter: 86%
Assuming that for this year the only anticipated change in sales from one month to the next is due to
this seasonal effect, what do these figures suggest for the likely number of units of this product sold:
a in the whole year?
b in each of the remaining three quarters of the year?

Copy and complete each of the following sets of seasonal data, filling in the spaces marked ‘?’. (Whilst
the average of the indices should be 100% do not worry if, due to rounding, this is not quite the case.)
5 Week Period of time Attendance Daily mean Attendance as % of

for the week week’s daily mean
1 Friday 2346 2766 84.82%
Saturday 3143 113.63%
Sunday 2809 101.55%
2 Friday 2572 ? ?
Saturday 3258 ?
Sunday 2957 100.96%
3 Friday 2987 ? ?
Saturday 3500 ?
Sunday 3206 99.23%
Friday Saturday Sunday

1st week 84.82% 113.63% 101.55%
2nd week ? ? 100.96%
3rd week ? ? 99.23%
Index (1 decimal place) ? ? 100.6%
6 Year Period of time Units sold Quarterly mean Units as % of year’s

for the year quarterly mean
1 1st quarter 54 000 52 650 102.56%
2nd quarter 63 200 120.04%
3rd quarter 45 200 85.85%
4th quarter 48 200 ?
2 1st quarter 48 500 ? 102.65%
2nd quarter 56 700 ?
3rd quarter 41 300 ?
3 1st quarter 51 800 ? 104.28%
2nd quarter 62 600 ?
3rd quarter 40 500 ?
1st quarter 2nd quarter 3rd quarter 4th quarter

1st year 102.56% 120.04% 85.85% ?
2nd year 102.65% ? ? ?
3rd year 104.28% ? ? ?
Seasonal index (1 decimal place) 103.2% ? ? ?

7 A local council commences a campaign to encourage people who need to contact the council for
various matters to do so via the council’s interactive website, thus avoiding the delays experienced
by people making phone or direct contact with the council’s help desk.
To monitor the success of this venture in
reducing the number of calls coming into
the help desk, and to be able to predict future
staffing needs, the council monitors the
number of non-internet phone calls made
to the help desk on each day of a three-week
period. The results are shown in the
table below.
The council wants to know appropriate
moving average figures for the three-week
itruts/moc.kcotSi
period and the seasonal index for each day.
With the assistance of a spreadsheet, construct
completed versions of the following tables.
n Week Day No. of calls 5-pt MA Calls as % of daily

average for week
1 Mon 258 – 107.59%
2 Tue 231 – 96.33%
3 One Wed 215 239.8
4 Thur 248 235.6
5 Fri 247 231.0
6 Mon 237
7 Tue 208
8 Two Wed 184
9 Thur 232
10 Fri 223
11 Mon 232
12 Tue 194
13 Three Wed 183
14 Thur 209 –
15 Fri 208 –
Week 1 Week 2 Week 3 Seasonal index

Monday 107.59%
Tuesday 96.33%
Wednesday
Thursday
Friday

Deseasonalising, or seasonally adjusting, the data
Now that we are able to quantify the effect a season has, we can remove this seasonal effect and present
the data deseasonalised or seasonally adjusted. Considering again the visitor numbers for the tourist
attraction situation encountered earlier:
Year Period of time Visitors (1000s)

We calculated earlier that the seasonal index for the 1st 4 months is 124%, or 1.24. Thus the figure of
45 000 visitors includes a 24% ‘seasonal increase’. To remove this 24% increase we divide the 45 000
by 1.24 to give 36 thousand, to the nearest thousand.
The seasonal index for the 2nd 4 months is 79%, or 0.79. Thus the figure of 26 000 visitors includes
a 21% ‘seasonal decrease’. To remove this 21% decrease we divide the 26 000 by 0.79 to give 33 000,
to the nearest thousand.
Continuing in this way gives the seasonally adjusted figures shown in the table below:
Year Period of time Visitors (1000s) Seasonally adjusted

visitor numbers (1000s)
2011 During 1st 4 months 45 36
During 2nd 4 months 26 33
During 3rd 4 months 34 35

Seasonally adjusted unemployment figures
• Why might the seasons of the year cause the number of people employed in agriculture to vary?
• Why might the seasons of the year cause the number of people employed in hospitality to vary?
• Why might the number of people registered as unemployed rise at the end of a school year?
• Why is it common practice for government statistical agencies to quote unemployment rates
as seasonally adjusted figures?
Making predictions
Let us suppose that a bank decides to operate a 7-days-a-week phone helpline to give advice regarding
new home loans and that the number of calls this helpline receives in the first three weeks of operation
are as shown in the table below.
Can we use these figures to predict the number of calls n Week Day No. of calls
this helpline is likely to receive on the Monday and
1 Mon 14
Saturday of week 4 (assuming any trends of the first
three weeks continue)? 2 Tue 13
3 Wed 14
If we plot the data, as shown below, it might appear that
4 1 Thur 15
any prediction would be difficult given the fluctuating
nature of the points. 5 Fri 22
6 Sat 28
Number of calls
50
7 Sun 34
8 Mon 20
45
9 Tue 17
40 10 Wed 18
35 11 2 Thur 20
12 Fri 27
30
13 Sat 36
25 14 Sun 44
20 15 Mon 25
16 Tue 21
15
17 Wed 19
10
18 3 Thur 23
5 19 Fri 34
M T W T F S S M T W T F S S M T W T F S S 20 Sat 44
2 4 6 8 10 12 14 16 18 20 n
21 Sun 51
However, if we add the 7-point moving average data to this graph, or if we plot the deseasonalised data,
both shown on the next page, we have straighter lines, showing less fluctuation. We could then, with more
confidence, use linear regression techniques to predict future values (although we know that such prediction
would involve extrapolation and so any predicted values would need to be viewed with caution).

Number of calls Number of calls
50 50
45 45
40 40
35 35
30 30
25 25
20 20
15 15
10 Deseasonalised data 10 7-point moving averages
5 5
M T W T F S S M T W T F S S M T W T F S S M T W T F S S M T W T F S S M T W T F S S
2 4 6 8 10 12 14 16 18 20 n 2 4 6 8 10 12 14 16 18 20 n
• With each plotted point on the line of moving averages being the average of seven points it is
perhaps to be expected that the moving average line shows less variation than the deseasonalised
data line. One ‘unusual value’ will have limited influence in changing the direction of a line
when it is but one of seven scores given equal weighting when determining the moving average.
Moving averages tend to remove seasonality and individuality. They smooth all variation, not just
seasonal variation.
• The deseasonalised data was formed using the following seasonal indices calculated using the
average percentage method explained earlier in the chapter.
Monday: 0.76 Tuesday: 0.66 Wednesday: 0.67 Thursday: 0.75
Friday: 1.08 Saturday 1.40 Sunday: 1.68
Now if we are to use either the moving average values, or the deseasonalised values, to determine a line
of best fit using linear regression techniques, any prediction we make using such a line of best fit will be
a predicted moving average value, or a predicted deseasonalised value. For these predictions to be ‘real-
life values’, rather than values that have been smoothed in some way, we will need to factor back in the
seasonal effect.
For example, if we use linear regression to determine a line of best fit for predicting D, the
deseasonalised data, given the value of n, the ‘day number’, where n = 1 corresponds to Monday of
week 1, we obtain the equation D = 0.70n + 18.0.
Thus: When n = 22 (i.e. Monday of week 4), predicted D = 33.4.
When n = 27 (i.e. Saturday of week 4), predicted D = 36.9.
Hence to ‘factor back in the seasonal effect’ the predictions will be as follows:
For Monday of week 4: Predicted number of calls = 33.4 × 0.76, i.e. approx. 25.
For Saturday of week 4: Predicted number of calls = 36.9 × 1.40, i.e. approx. 52.

Note: Remember that not all data can be suitably modelled using linear regression. Just because
smoothing techniques such as deseasonalising the data or finding the moving averages have been
carried out we cannot assume that the resulting data is necessarily showing a linear pattern.
Recall for example the wool production data encountered earlier in this chapter, for which the
moving average plot is shown below.
Production (tonnes)
1 000000
800000
600000
400000
200000
Year
0291
0391
0491
0591
0691
0791
0891
0991
0002
0102
As the graph indicates, applying linear regression techniques to the whole data set would not
be appropriate.
Production (tonnes)
1 000000
800000
600000
400000
200000
Year
0291
0391
0491
0591
0691
0791
0891
0991
0002
0102
The straight line given by using linear regression is shown above and is not a very good fit for
the data.
Remember:
Real data
Deseasonalised data (or seasonally adjusted data) =
Seasonal index
and so
Real data = Deseasonalised data × Seasonal index

Note: In this unit we are using a multiplicative model for determining the seasonal indices. Hence the
statements at the base of the previous page are true. In some cases, though not in this unit,
an additive model could be used and in such cases seasonal adjustment occurs by addition or
subtraction, not by multiplication.
Exercise 2D
1 A company achieves profits of $25 400 in June of one year. If the June profit figures for this
company have a seasonal index of 114% what would be the deseasonalised, or seasonally adjusted,
profit for June? (Give your answer to the nearest $100.)
2 From an analysis of sales from earlier years, a company statistician determines that the number
of units of a particular product the company sells follows a seasonal pattern with the following
quarterly indices:
First quarter 0.65 Second quarter 0.76
Third quarter 0.82 Fourth quarter 1.77
Deseasonalise a figure of 132 000 units sold in the second quarter, giving your answer to the
nearest 100 units.
3 a It is announced that a particular region had 65 470 people unemployed in January ‘seasonally
adjusted’. If the seasonal index for January is 106.2% what was the ‘real number’ unemployed
for this region in January? (Give your answer to the nearest ten people.)
b In May of the same year the region had 67 140 people unemployed. If the seasonal index for
May is 96% what would the real figure of 67 140 be reported as after ‘seasonal adjustment’?
(Again give your answer to the nearest ten people.)
4 Projecting forward using smoothed data a company analyst finds that the predicted sales figure for
January will be 12 500 units. However this has not allowed for the January seasonal effect. If the
seasonal index for January is 1.18 what does this predict for the number of units sold in January?
5 Projecting forward using smoothed data the statistically astute owner of a number of orchards
predicts that the orchards will produce 16 000 kg of a particular fruit this summer, but he knows
that this prediction has not yet allowed for the seasonal effect of summer. If the summer seasonal
index is 1.48, what does this suggest the actual weight of this fruit from the orchards will be
this summer?
noillatsder/moc.kcotSi

6 Let us suppose that the four-monthly totals for short-stay visitors arriving in a particular country,
collected over a four-year period, are as shown in the table below.
a Calculate the seasonal indices for each of the seasons Jan → Apr, May → Aug, Sept → Dec.
(Give indices as percentages correct to 2 decimal places, e.g. 105.72%.)
b Deseasonalise the numbers in the arrivals column.
n Year Months Arrivals (millions)

1 Jan → Apr 1.83
2 1 May → Aug 1.29
3 Sept → Dec 1.98
4 Jan → Apr 1.87
5 2 May → Aug 1.41
6 Sept → Dec 2.06
7 Jan → Apr 1.93
8 3 May → Aug 1.28
9 Sept → Dec 2.16
10 Jan → Apr 2.18
11 4 May → Aug 1.52
12 Sept → Dec 2.33
7 A show is held each year over a long weekend, starting on the Friday and finishing on the public
holiday Monday. The attendance figures for four consecutive years are shown in the table below.
a Calculate the seasonal index for each of the four days. (Give your answers as a decimal and
correct to 4 decimal places, e.g. 1.1234.)
b Seasonally adjust the numbers in the attendance column. (Answer to the nearest 10.)
Year Day Attendance

Friday 9 200
Saturday 14 840
1
Sunday 16 260
Monday 16 850
Friday 9 810
Saturday 15 790
2
Sunday 17 270
Monday 17 800
Friday 9 450
Saturday 15 220
3
Sunday 17 200
Monday 16 810
Friday 9 840
Saturday 16 900
4
Sunday 18 040
Monday 16 750

8 The table below shows the evening takings for a new take-away fish and chip shop, each evening
for the first three weeks of operation.
t Week Day Takings Deseasonalised

weekly takings (D)
1 Monday $537
2 Tuesday $618
3 Wednesday $372
4 1 Thursday $770
5 Friday $937
6 Saturday $915
7 Sunday $492
8 Monday $605
9 Tuesday $730
10 Wednesday $410
11 2 Thursday $880
12 Friday $1085
13 Saturday $1003
14 Sunday $495
15 Monday $705
16 Tuesday $804
17 Wednesday $420
18 3 Thursday $840
19 Friday $1240
20 Saturday $1270
21 Sunday $531
a Given that the seasonal indices are as shown below, calculate the entries for the deseasonalised
weekly takings column, to the nearest dollar.
Monday Tuesday Wednesday Thursday Friday Saturday Sunday
82.42% 96.07% 53.94% 111.87% 145.52% 141.94% 68.24%
b Use a graphic calculator or computer spreadsheet to view a graph showing the deseasonalised
figures, D, on the vertical axis plotted against t on the horizontal axis (t is as shown in the
above table) to confirm that for the variables t and D linear regression would be appropriate.
c Use t and the deseasonalised data, D, to determine the equation of the regression line D = A + Bt.
d Use the regression equation from part c and the seasonal indices to predict the takings for
Monday, Wednesday and Saturday of week 4.

9 The number of students absent from a school on each day for the first four full weeks of a term are
shown in the table below.
n Week Day No. absent 5-pt MA (M) Seasonally

adjusted absences
1 Mon 35 – 32
2 Tue 25 –
3 One Wed 24 30.4
4 Thur 28 30.8
5 Fri 40 31.8 30
6 Mon 37 32.4 34
7 Tue 30 32.8
8 Two Wed 27 34.4
9 Thur 30 36.2
10 Fri 48 36.6 36
11 Mon 46 37.8 42
12 Tue 32 38.8
13 Three Wed 33 40.2
14 Thur 35 40.6
15 Fri 55 42.6 42
16 Mon 48 44.0 43
17 Tue 42 45.8
18 Four Wed 40 46.0
19 Thur 44 –
20 Fri 56 – 42
a Use a graphic calculator or computer spreadsheet to view the 5-point moving averages
(M) on the vertical axis plotted against n on the horizontal axis, to confirm that for the
variables n and M, linear regression would be appropriate.
b Use n and the 5-point moving averages, M, to determine the equation of the regression line
M = An + B.
c Create a completed version of the following to calculate the seasonal indices.
Monday Tuesday Wednesday Thursday Friday

Week 1 1.151 32 1.315 79
Week 2 1.075 58 1.395 35
Week 3 1.144 28 1.368 16
Week 4 1.043 48 1.217 39
Mean (4 dp) 1.103 7 1.324 2
d Calculate the missing values in the final column in the first table
e Use this regression equation M = An + B and the seasonal indices to predict the absences for
each day of week five.
f Suggest some real-life events that could make these predictions unreliable.

10 Let us suppose that the number of rescues carried out by a surf rescue group in each two months
of its six-months season, over four seasons are as follows:
n Season Months Rescues Season’s 2-month mean

1 Oct/Nov 46
2 1 Dec/Jan 111 82
3 Feb/Mar 89
4 Oct/Nov 36
5 2 Dec/Jan 104 76
6 Feb/Mar 88
7 Oct/Nov 36
8 3 Dec/Jan 107 70
9 Feb/Mar 67
10 Oct/Nov 34
11 4 Dec/Jan 91 64
12 Feb/Mar 67
a Create a completed version of the following:
Rescues as a percent of season’s 2 month mean

(as percentages and correct to 2 decimal places)
Oct/Nov Dec/Jan Feb/Mar
Season 1 56.10% 135.37%
Season 2
Season 3
Season 4
Seasonal indices
52%
(nearest percent)
b Produce deseasonalised figures, D, for each of the two-month periods of seasons 1 to 4.

(Give the D values to the nearest integer.)
c Use a graphic calculator or computer spreadsheet to view a graph showing the deseasonalised
figures, D, on the vertical axis plotted against n on the horizontal axis (n is as shown in the
initial table), to confirm that for the variables n and D linear regression would be appropriate.
Hence determine the equation of the regression line
D = A + Bn .
d Use the regression equation from part b and the seasonal indices to predict the number
of rescues for each two-month period of season 5.

INVESTIGATION
Internet websites for government agencies such as

The Australian Bureau of Statistics
and
The Bureau of Meteorology
have a wealth of time series data available.
For example
Monthly numbers of house* building commencements approved.
(*Referred to as ‘dwelling units’.)
Quarterly consumer sales figures.
Monthly maximum temperatures.
Etc.
Investigate.
Miscellaneous exercise two

This miscellaneous exercise may include questions involving the work of this chapter, the
work of any previous chapters, and the ideas mentioned in the Preliminary work section at the
beginning of the book.
1 Increase $250 by 10%, then decrease your answer by 10%.
2 The rule for converting a Celsius temperature (°C) to the corresponding Fahrenheit temperature
(°F) is
9
F = C + 32.
5
What is the value of the correlation coefficient rCF?
3 What does it mean if the seasonal index for February is 0.87 (or 87%)?
4 Deseasonalise the following raw data given the seasonal indices are as stated.
Season Spring Summer Autumn Winter

Raw data 1400 1940 1520 750
Seasonal index 1.40 1.25 0.95 0.4

5 Starting Increase by 10% Decrease by 10%
amount → →
Increase by 20%
↓
Decrease by 80% Increase by 50%

← ←
Increase by 40%
↓
Increase by 50% Decrease by 60%

$449 064
→ →
Find the starting amount.
6 Counting the year 1998 as zero, the year of manufacture and the asking price of a certain make
and model of vehicle advertised in a newspaper one day in 2014 were as follows:
Year No. 5 13 6 4 4 15 6
Asking price ($) 19 980 44 500 17 000 13 500 11 400 57 300 20 500
Year No. 11 10 7 12 10 6
Asking price ($) 39 000 32 900 25 000 35 700 36 800 25 500
Display these figures graphically with t, the year number in the above table, on the horizontal axis
and p, the asking price, on the vertical axis.
Comment on any relationships shown on your graph.
Use your calculator or a spreadsheet to determine the equation of the line of best fit in the form
p = at + b and use this to predict the likely price in 2014 for a 2007 vehicle of this make and model.
7 Using sales data from previous years a company determines that the quarterly revenues achieved
from sales of its products seem to follow a reasonably steady seasonal pattern with the following
seasonal indices:
1st quarter 0.85 2nd quarter 1.25
3rd quarter 1.12 4th quarter 0.78
The following year the company finds that the quarterly revenue figures are as follows:
1st quarter $752 000 2nd quarter $1 138 000
3rd quarter $1 176 000 4th quarter $694 000
a If the above information is all we know about the company, why are the quarterly revenues
shown above a surprise?
b If we were informed that the company had a big advertising campaign at the start of the
3rd quarter might this make the revenue figures less surprising?

8 The table below shows the sales figures for a particular item on a quarter year basis, for three years.
a Create completed versions of the following tables using the average percentage method to
determine seasonal indices.
t Year Quarter Number sold Number sold as Number sold

in quarter percentage of quarterly seasonally
mean for the year adjusted
1 One 1st 1637 106.99% 1514
2 2nd 1489
3 3rd 1244
4 4th 1750
5 Two 1st 1405
6 2nd 1241
7 3rd 1012
8 4th 1538
9 Three 1st 1253
10 2nd 1121
11 3rd 852
12 4th 1362
Year 1 Year 2 Year 3 Seasonal index

1st quarter 106.99% 108.13%
2nd quarter
3rd quarter
4th quarter
b Use a graphic calculator or computer to view the graph of the seasonally adjusted figures,
which we will call S , plotted on the vertical axis, against t on the horizontal axis, and confirm
that applying linear regression techniques is appropriate.
c Use t and the seasonally adjusted sales figures to determine the equation of the regression line
S = at + b.
d Use this regression equation and the seasonal indices to predict the number of the particular
item sold for each quarter of year 4.

3.
Finance I: Saving
and borrowing
• Revision of simple interest, compound interest and
recursive rules
• Effective annual interest rate
• Initial deposit plus regular investments
• Compounding isn’t only about investing
• Loans with regular repayments
• What price house can these people afford?
• Miscellaneous exercise three
Revision of simple interest, compound interest WS
and recursive rules Simple interest
The Preliminary work at the beginning of this book reminded you of simple interest, compound interest
and sequences defined by giving the first term and a recursive rule. WS
The next examples and exercise 3A that follows should serve as a further reminder.
Applications of
Note: The simple interest, $I, earned when $P is invested for T years in an account paying
simple interest
PRT
R% per annum simple interest is given by I =
WS
100
If instead we use R in decimal form we use I = PRT. Using spreadsheets
to calculate interest
For example if the interest rate is 6% the first formula would use R = 6 but the second would
use R = 0.06. WS
Compound interest
EXAMPLE 1
How much interest would be earned in 4 years on an investment of $5000 at 6% per annum
simple interest?
Solution
Using I = PRT with P = 5000, R = 0.06 and T = 4
I = 5000 × 0.06 × 4
= 1200
The interest earned will be $1200.
EXAMPLE 2
What annual rate of simple interest will see an investment of $8000 earn $3600 simple interest
in 6 years?
Solution
Using I = PRT with I = 3600, P = 8000 and T=6
3600 = 8000 × R × 6
3600 = 48 000R
3600
R =
48 000
= 0.075
The required rate of simple interest is 7.5%.
ISBN 9780170395069 3. Finance I: Saving and borrowing 53

EXAMPLE 3
A simple interest arrangement of 5.4% per annum sees an initial investment grow to $11 121.60 in
6 years. What was the initial investment?
Solution
Using I = PRT with P + I = 11 121.60, R = 0.054 and T = 6
P+I = P + PRT
11 121.60 = P + P × 0.054 × 6
11 121.60 = P + 0.324P
11 121.60 = 1.324P
P = 8400
The initial investment was $8400.
EXAMPLE 4
$25 000 is invested for 4 years with an annual compound interest rate of 6%. Find the amount this
account is worth at the end of the 4 years if the compounding occurs
a annually, b quarterly, c daily.
Solution
a Compounding annually.
There will be 4 compoundings in the 4 years, each of 6%.
Value after 4 years = $25 000 ×1.06 × 1.06 × 1.06 × 1.06
= $25 000 × 1.064
= $31 561.92 to the nearest cent.
b Compounding quarterly.
There will be 4 compoundings each year, each of 1.5% (= 6% ÷ 4).
I.e. 16 in the 4 years
c Compounding daily.
There will be 365 compoundings each year, each of (6% ÷ 365).
I.e. 1460 (= 365 × 4) in the 4 years
365 × 4
 0.06 
Value after 4 years = $25 000 ×  1 + 
 365 
Note: • For convenience, in this book, we will ignore leap years and assume that all years have
365 days.
• Prior to the ready availability of computers and calculators, some banking calculations,
again for convenience, were based on a concept called a banker’s year. This concept takes
a year as being 360 days and consisting of twelve equal months each of thirty days. This
concept is mentioned here for information only, it will not be used in this course.

EXAMPLE 5
At an 8% annual compound interest rate, with compounding every six months, how many years
would it take for an initial investment of $2500 to grow to $3700?
Solution
There will be 2 compoundings each year, each of 4%.
If the account runs for T years:
$3700 = $2500 × 1.042T
Using the solve facility of some calculators gives T = 4.998 (rounded to 3 decimal places).
It would take 5 years.
Consider an investment of $8000 at 5% per annum simple interest. The annual value of this investment
forms the sequence:
$8000 $8400 $8800 $9200 $9600 …
i.e. an arithmetic progression with first term $8000 and common difference $400.
Recursive rule: T1 = $8000, Tn 1 = Tn + $400.
+
Or, if we want Tn to represent the value after n years, T0 = $8000, Tn + 1 = Tn + $400.

Had the investment been at 5% compound interest, with interest compounded annually, the sequence
would be:
$8000 $8400 $8820 $9261 $9724.05 …
i.e. a geometric progression with first term $8000 and common ratio 1.05.
Recursive rule: T1 = $8000, Tn 1 = 1.05 × Tn.
+
Or, if we want Tn to represent the value after n years, T0 = $8000, Tn + 1 = 1.05 × Tn.
In this way recursive rules can be useful for simple and compound interest situations.
an+1 = a n + 400 an+1 = 1.05•a n

a1 = 8000 a1 = 8000
bn+1 = bn+1 =
b1 = 0 b1 = 0
cn+1 = cn+1 =
c1 = 0 c1 = 0
n an n an
1 8000 1 8000
2 8400 2 8400
3 8800 3 8820
4 9200 4 9261
5 9600 5 9724.1
9200 9724.05

Exercise 3A
Simple interest
1 What will be the value after 6 years of $2000 invested at 6% per annum simple interest?
2 How long will it take for $800, invested in an account paying 7.5% per annum simple interest,
to become $1100?
3 What annual rate of simple interest is needed to see an investment of $5000 become $7500 in
4 years?
4 How much must have been invested in an account paying 8% simple interest if after 6 years the
total interest earned was $1152?
5 What annual rate of simple interest is needed to see an initial investment of $6600 become $7007
in 10 months?
6 A simple interest arrangement of 6.5% per annum sees an initial investment grow to $10 703 in
6 years. What was the initial investment?
Compound interest
7 $4000 is invested at 8% per annum compounded annually. Determine the value of this investment
after 6 years.
8 $500 is left in an account paying 6% per annum compounded six-monthly. How much will this
account be worth 50 years later?
9 How much interest is earned on $50 000 invested for 4 years if the interest rate is 6% per annum
compounded annually?
How much more interest would be earned in the 4 years if instead the interest is
compounded monthly?
10 How much interest is earned on $400 000 invested for ten years if the interest rate is
8% per annum compounded annually?
How much more interest would be earned in the ten years if instead the interest is
compounded daily?
11 How much must be invested now into an account paying 7.5% compound interest, compounded
annually, for the account to be worth $10 000 in five years?
12 If an amount is invested at 6% per annum compound interest, with interest compounded quarterly,
how long would it take for the initial investment to double in value?
13 At a 9% annual compound interest rate, with compounding every six months, how long would it
take for an initial investment of $5000 to first exceed $12 000?

Recursive rules
Determine the first five terms and, with the help of a calculator that is able to handle recursive rules
if necessary, the fifteenth term, of each of the following recursively defined sequences.
14 Tn + 1 = Tn + 3, T1 = 5.
15 Tn + 1 = Tn - 3, T1 = 5.
16 un + 1 = un + 5, u1 = –10.
17 an = an - 1 + 2.5, a1 = 12.5.
18 Tn + 1 = 2 × Tn, T1 = 0.25.
19 Tn + 1 = 1.5 × T n, T1 = 20 480.
20 an + 1 = 0.5 × an, a1 = 2 621 440.
21 un + 1 = 0.5 × un, u1 = 230. (Give answers as powers of 2.)
22 Tn + 1 = 2 × Tn + 3, T1 = 5.
23 Tn + 1 = 1.2 × T n + $100. T1 = $1000. (Write T15 to the nearest cent.)
24 Tn + 1 = 1.05 × T n - $250. T1 = $5000.
25 Tn = an + un where an + 1 = an + 6, a1 = 5,
and un + 1 = un + 5, u1 = 7.
26 Tn = an × un where an + 1 = 2 × an, a1 = 5,
and un + 1 = 0.5 × un, u1 = 163 840.
27 Tn + 2 = Tn + 1 + Tn, T1 = 1, T2 = 1.
28 Tn + 2 = 2 × Tn + 1 + Tn, T1 = 2, T2 = 3.
29 Tn + 1 = (–1)nTn, T1 = 5.
30 Tn + 1 = (–1)n × T n + 2, T1 = 5.

Effective annual interest rate
Consider $100 invested in an account paying 6% per annum compound interest.
After one year, with compounding …
occurring annually, the account will be worth $100 × 1.06
= $106
occurring six monthly, the account will be worth $100 × 1.032
= $106.09
occurring quarterly, the account will be worth $100 × 1.0154
= $106.136 (correct to 3 decimal places)
occurring monthly, the account will be worth $100 × 1.00512
365
 0.06 
occurring daily, the account will be worth $100 ×  1 + 
 365 
Thus whilst 6% per annum returns $6 interest in the year when compounded annually, more frequent
compoundings sees our 6% interest rate return more than $6 on $100 invested, i.e. more than 6%.
The 6% is said to be the nominal annual interest rate but to make comparison between the effect
of the different compounding periods we call the 6.09%, 6.136%, 6.168% and 6.183% returns shown
above the effective annual interest rates.
Thus for a nominal compound interest rate of 6% per annum the effective annual interest rate for
six-monthly compounding is 6.09%. The effective annual interest rate for quarterly compounding is
6.136% (correct to three decimal places), the effective annual interest rate for monthly compounding
is 6.168% (correct to three decimal places) etc.
This effective annual interest rate can be determined by consideration of the compound interest earned
in one year by an investment of $100, as shown above, or by use of the following formula:
For a nominal interest rate of i per annum and n compounding periods per year,
 i n
Effective annual interest rate = 1+ - 1.

 n 
This is sometimes written as
i n
i effective =

1+ - 1.

 n 
Readers should check that with suitable rounding this formula does indeed give 0.0609 (i.e. 6.09%),
0.061 36 (i.e. 6.136%) and 0.061 68 (i.e. 6.168%), for i = 0.06 and values for n of 2, 4 and 12
respectively, as quoted above.

EXAMPLE 6
Find the effective annual interest rate for a nominal interest rate of 8% per annum with
compounding occurring quarterly.
Solution
By considering $100. By the formula.
8% per annum is 2% per quarter. i n
Using ieffective =

1+ -1
After 1 year a $100 investment becomes


 n 
with i 0.08
$100 × 1.024
=
and n = 4.
= $108.243 (correct to 3 decimal places) 4
0.08 
ieffective =


1+ 
-1
 4 
= 0.082 432 16
The effective annual interest rate is 8.243% (correct to 3 decimal places).
Note: If a question simply refers to an annual interest rate of, say, 8%, it should be assumed
that it is the nominal annual rate that is being quoted.
Exercise 3B
1 Copy and complete the following table. (1 year = 12 months = 52 weeks = 365 days.)
Compounding Nominal annual Effective annual

frequency interest rate (%) interest rate (%)
Annual 4%
Six-monthly 4%
Quarterly 4%
Monthly 4%
Weekly 4%
Daily 4%

Annual 8%
Six-monthly 8%
Quarterly 8%
Monthly 8%
Weekly 8%
Daily 8%
3 Looking at your answers for the effective annual interest rates in questions 1 and 2, are the rates
for the 8% situation simply twice those of the 4% situation?
If not, why not?

INVESTIGATE
Does your calculator have a nominal interest ↔ effective interest facility?

Are there any ‘effective interest rate calculators’ on the internet?
INVESTIGATE
Another type of interest rate, which at the time of writing is not specifically mentioned in the
syllabus for this unit, is the real interest rate.
Just as the effective rate is the rate once the effects of compounding have been allowed for, the
real interest rate is the rate once inflation has been allowed for.
Initial deposit plus regular investments

The investments considered so far in this chapter have all involved a ‘one-off’ investment deposited
in an account and left to earn interest until some later date. In practice, many savings schemes involve
an initial amount being invested followed by further deposits into the account, perhaps on a regular
basis. Consider for example an initial investment of $5000 into an account paying compound interest of
6% per annum, compounded annually, with an extra $500 deposited into the account each year thereafter.
Initial deposit = $5000
Amount in account at end of 1 year = $5000 × 1.06 + $500
= $5800
Amount in account at end of 2 years = $5800 × 1.06 + $500
= $6648
Amount in account at end of 3 years = $6648 × 1.06 + $500
= $7546.88
With T1 = $5000, T2 = $5800, T3 = $6648, T4 = $7546.88 etc, what we have here is a sequence that can
be defined recursively by the rule:
Tn + 1 = 1.06 × Tn + $500. T1 = $5000.
Or, if we want the end of year number to be the same as the term number we use
Tn + 1 = 1.06 × Tn + $500. T0 = $5000.
Then T 1 will give the value at the end of year one, T2 will give the value at the end of year two, T3 will
give the value at the end of year three, etc.
The progress of the investment can then be easily viewed using a calculator with an ability to display
the terms of a recursively defined sequence, or using a computer spreadsheet – see the displays on the
next page.

an+1 = 1.06•an + 500 A B C D
a1 = 5000 1 Initial investment $5,000.00
bn+1 = 2 Interest rate per yr (%) 6.00
b1 = 0 3 Annual deposit $500.00
4 Balance at end of year 1 $5,800.00
cn+1 =
5 2 $6,648.00
c1 = 0
6 3 $7,546.88
n an 7 4 $8,499.69
8 5 $9,509.67
4 7546.9
9 6 $10,580.25
5 8499.7
6 9509.7 10 7 $11,715.07
7 10580. 11 8 $12,917.97
8 11715. 12 9 $14,193.05
11715.0701198851 13 10 $15,544.64
• Remember that in the display above left, with the first term being the value after zero years, the
8th term will be the value of the investment after 7 years.
Create a spreadsheet like the one shown above right. Try to make your spreadsheet adaptable to
other situations of this type such that by simply changing the amounts in cells D1, D2 and D3 the
annual balances are automatically recalculated.
EXAMPLE 7
At the beginning of one month, $1000 is invested into an account paying interest at 7.5% per
annum, compounded monthly, and an extra $100 is invested at the end of that first month and
the end of every month thereafter. How much is the account worth at the end of the month two
years later, just after the $100 deposit for that month is made? (For the purposes of this question
assume that a year consists of 12 equal months.)
Solution
Initial value = $1000
 0.075 
Value at end of 1st month = $1000 ×  1 + 
+ $100 i.e. $1106.25
 12 
Value at end of 2nd month = $1106.25 × 1.006 25 + $100
Thus with T n the value of the investment at the end of the nth month
Tn + 1 = 1.006 25 × Tn + $100. T0 = $1000.
Either using the facility of some calculators to display the terms of recursively defined sequences,
or by using your version of the spreadsheet shown above right, with appropriate entries suitably
changed (see next page):
Value at end of 24th month = T24
= $3741.96 (nearest cent).
Hence two years after the initial investment, and just after the $100 deposit for that month has
been made, the account will be worth $3741.96.

A B C D
1 Initial investment $1,000.00
2 Interest rate per month (%) 0.625
3 Monthly deposit $100.00
4 Balance at end of month 1 $1,106.25
5 2 $1,213.16
6 3 $1,320.75
an+1 = 1.00625•an + 100 7 4 $1,429.00
a0 = 1000 8 5 $1,537.93
bn+1 = 9 6 $1,647.54
b0 = 0 10 7 $1,757.84
11 8 $1,868.83
cn+1 =
12 9 $1,980.51
c0 = 0
13 10 $2,092.89
14 11 $2,205.97
n an
15 12 $2,319.75
20 3256.0 16 13 $2,434.25
21 3376.4
17 14 $2,549.47
22 3497.5
23 3619.3 18 15 $2,665.40
24 3742.0 19 16 $2,782.06
20 17 $2,899.45
3741.9643079925
21 18 $3,017.57
22 19 $3,136.43
23 20 $3,256.03
24 21 $3,376.38
25 22 $3,497.48
26 23 $3,619.34
27 24 $3,741.96
Some calculators have built-in financial programs that allow some standard financial calculations to be
performed automatically, given appropriate information.
The display on the right shows typical information that would be
displayed for the situation just encountered. Compound Interest
N 24
The N in the display is the number of investment periods, in this
case 24 months. I(%) 7.5
I (%) is the annual interest rate as a percentage, in this case 7.5. PV –1000
The initial investment of $1000 is shown as PV, for present value. Pmt –100
This is shown as a negative amount because from the investor’s FV 3741.964308
point of view it is money being deposited (outgoing), not money
PpY 12
being received.
CpY 12
The additional investment per month (Pmt) is $100, and again from
the investor’s point of view this is shown as a negative (outgoing).
Payments per year and compoundings per year are each 12.
Provided the program has the correct settings, e.g. the payment date set as being at the end of each
compounding period, not at the beginning, when asked to determine the final value ( FV) the calculator
returns the final value as $3741.96, as before.

Exercise 3C
For questions 1 and 2 use recursion. For 3, 4 and 5 do each question twice, once using recursion
and once using a calculator with financial capability.
1 $4000 is invested into an account paying interest at 8% per annum, compounded annually, and
an extra $200 is invested at the end of each 12 months. Thus:
Amount in account at end of 1 year = $4000 × 1.08 + $200 ← T1
Amount in account at end of 2 years = ($4000 × 1.08 + $200) × 1.08 + $200 ← T2
Express Tn 1 in terms of T n and determine (to the nearest cent) the amount in the account at the
+
end of ten years, after the $200 for that year has been added.
2 $5000 is used to open an investment account paying interest at 6% per annum, compounded
monthly, and an extra $100 is invested at the end of each one-month period thereafter. Thus,
the value of the account progresses as follows:
Initially = $5000 ← T0
At the end of 1 month = $5000 × 1.005 + $100 ← T1
At the end of 2 months = ($5000 × 1.005 + $100) × 1.005 + $100 ← T2
a Express Tn + 1 in terms of Tn.
b The account is closed at the end of three years, without the $100 for the end of the final
month being made. Determine the final value of the account.
3 At the beginning of 2014 Tenielle opens a savings account by depositing $2500 into an account
paying 8% per annum, compounded annually. Tenielle plans to follow that initial deposit with
further deposits each of $1000 made at the end of every period of one year thereafter.
Assuming Tenielle keeps to her plans with regards to the further deposits, how much is the
account worth at the beginning of 2025 (i.e. just after the $1000 deposit for the end of 2024
has been made)?
4 At the beginning of 2014 Sanchez opens a savings account by depositing $500 into an account
paying 7.8% per annum, compounded annually. Sanchez plans to follow that initial deposit with
further deposits each of $500 made at the end of every period of one year thereafter.
Assuming Sanchez keeps to his plans with regards to the further deposits, how much is the account
worth at the end of 2020, i.e. just before the $500 deposit is made?
5 At the beginning of one month $5000 is invested into an account paying interest at 9% per
annum, compounded monthly, and an extra $200 is invested at the end of that month and the
end of every month thereafter. How much is the account worth at the end of the month three
years later, just after the $200 deposit for that month?
6 $5000 is used to open an account paying interest that is compounded annually, and an additional
$1000 is to be added to the account at the end of every one-year period thereafter. Using a
spreadsheet, or a calculator able to display the terms of a recursively defined sequence, or financial
programs available on some calculators or internet sites, find the annual interest rate required for
this account to be worth
a $10 000 after 3 years (including the final $1000 payment),
b $100 000 after 25 years (including the final $1000 payment).

Compounding isn’t only about investing
In a financial context, compounding is not only applied when we invest money, it is also applied
to the interest we owe on borrowed money. For example, suppose you borrow $50 000 from a bank
or financial institution. You will be charged interest on this loan. Let us suppose that the interest is
charged at the rate of 6% per annum nominal rate, compounded annually.
After one year you will owe $50 000 × 1.06 i.e. $53 000
After two years you will owe $50 000 × 1.062 i.e. $56 180
After three years you will owe $50 000 × 1.063 i.e. $59 550.80 etc.
If instead the compounding occurred monthly:
After one year you will owe $50 000 × 1.00512 i.e. $53 083.89
After two years you will owe $50 000 × 1.00524 i.e. $56 357.99
After three years you will owe $50 000 × 1.00536 i.e. $59 834.03 etc.
Similar repeated multiplication on an annual basis can occur when the value of an asset is depreciated
each year.
Suppose a company purchases a new machine worth $100 000. One year later the machine is no longer
new so its value will have depreciated. One method used for determining the value at the end of each
year is to depreciate the value of the asset by a certain percentage of its value at the beginning of that
year. Thus if the machine initially worth $100 000 is depreciated at an annual rate of 12% its value in
later years will be as follows:
Value of the machine after 1 year = $100 000 × 0.88 i.e. $88 000
Value of the machine after 2 years = $100 000 × 0.882 i.e. $77 000 (nearest $1000)
Value of the machine after 3 years = $100 000 × 0.883 i.e. $68 000 (nearest $1000)
etc.
You may recall from Unit 3 of Mathematics Applications that the above form of depreciation, called the
reducing balance method (or fixed percentage method or diminishing value method) is not the only
method for calculating depreciation.
We also have: the flat rate, or straight line, method, in which the depreciation is a fixed amount
each year,
and the unit cost method, in which the depreciation is related to the units of production
the machine has produced.
For example suppose a machine has an initial value of $500 000.
Consider the following three depreciation schedules:
Flat rate depreciation of $50 000 per year
Initial value: $500 000
Value after 1 year: $450 000 Depreciation in 1st year: $50 000
Value after 2 years: $400 000 Depreciation in 2nd year: $50 000
Value after 3 years: $350 000 Depreciation in 3rd year: $50 000
Value after 4 years: $300 000 Depreciation in 4th year: $50 000 etc.

Reducing balance depreciation of 10% per year
Value after 1 year: $500 000 × 0.9 = $450 000 Depreciation in 1st year: $50 000
Value after 2 years: $450 000 × 0.9 = $405 000 Depreciation in 2nd year: $45 000
Value after 3 years: $405 000 × 0.9 = $364 500 Depreciation in 3rd year: $40 500
Value after 4 years: $364 500 × 0.9 = $328 050 Depreciation in 4th year: $36 450 etc.
Unit cost depreciation of $60 000 per 100 000 units of production
Value after 100 000 units of production: $440 000
Value after 200 000 units of production: $380 000
Value after 300 000 units of production: $320 000 etc.
Exercise 3D
1 An asset has an initial value of $220 000. Determine its value at the end of each of the first three
years using a flat rate depreciation of $25 000 per year,
b reducing balance depreciation of 15% per year.
2 An asset has an initial value of $80 000. By how much does the asset depreciate in each of its first
three years if the depreciation method used is a flat rate depreciation of $10 000 per year?
b reducing balance depreciation of 20% per year?
3 A printing machine initially worth $30 000 is expected to be worth just $2000 after it has been
used to produce one million copies.
Using the unit cost depreciation method what will be the value of the machine after it has produced
a 250 000 copies? b 500 000 copies? c 750 000 copies?
4 Some calculators with financial capabilities, and some online financial calculators, can work out
depreciation schedules. Investigate.
Loans with regular repayments

Rather like the investments that involved a regular contribution, people
repoortrepuS/moc.kcotsrettuhS
often make regular contributions to reduce the balance of a loan.
For example, suppose Jim wants to buy a motor scooter costing $4000.
Further suppose that Jim borrows the $4000 from a bank, that he
is charged 1% interest per month on the remaining balance and
he manages to repay $100 at the end of every period of one month.
Initially he owes $4000
After 1 month he will owe: $4000 × 1.01 - $100 = $3940
After 2 months he will owe: $3940 × 1.01 - $100 = $3879.40
After 3 months he will owe: $3879.40 × 1.01 - $100 = $3818.194 etc.,
the value of the loan reducing each month because the amount repaid exceeds the interest charged.
Such an arrangement is referred to as a reducing balance loan.

WS EXAMPLE 8
Reducing balance
loan: spreadsheet Johan takes out a loan of $7000. Interest of 9.4% per annum is added to the loan annually and
Johan repays $800 at the end of each one-year period. How much will Johan still owe immediately
after he makes the $800 repayment at the end of year 8?
Solution
Initial amount owed: $7000
Amount owed after 1 year: $7000 × 1.094 - $800 = $6858
Amount owed after 2 years: $6858 × 1.094 - $800 = $6702.65 (nearest cent)
Amount owed after 3 years: $6702.65 × 1.094 - $800 = $6532.70 (nearest cent)
We have Tn + 1 = Tn × 1.094 - $800,
and T0 = $7000
Using the recursion capability of some calculators, or a spreadsheet:
an+1 = 1.094•an – 800 A B C D

a0 = 7000 1 Amount borrowed $7,000.00
bn+1 = 2 Interest rate 9.40
b0 = 0 3 Annual repayments $800.00
4 Balance at end of year 1 $6,858.00
cn+1 =
5 2 $6,702.65
c0 = 0
6 3 $6,532.70
n an 7 4 $6,346.78
8 5 $6,143.37
4 6346.8
5 6143.4 9 6 $5,920.85
6 5920.8 10 7 $5,677.41
7 5677.4 11 8 $5,411.09
8 5411.1 12
5411.08529854263 13
The amount still owed at the end of year 8 is $5411.09.
Note: • Continuing down either of the previous tables

allows us to see when the loan will be reduced n an
to zero. In this case the payment at the end of 16 2150.9
the twentieth year would pay off the loan (and 17 1553.1
18 899.10
this final payment would only need to be
19 183.62
approximately $200, not $800). 20 –599.1
• By redefining the recursive rule using a different –599.121602634385
repayment amount, or by inserting a different
repayment amount into the appropriate cell of the
spreadsheet, we can investigate how much the annual repayment needs to be in order to pay
off the loan in the eight years. Investigate this for yourself and confirm that in this case the
required amount is $1283.58 (with the final repayment being a few cents more).

Once again we could use the financial programs available on some
calculators and internet websites that allow the remaining balance Compound Interest WS
of a loan after a number of repayments, or the repayment required N 8

Using finance solvers
to pay off a loan in a particular time etc., to be determined. The I(%) 9.4
for compound interest
loans
display on the right shows the $1283.58 mentioned in the previous
dot point (i.e. the regular repayment needed to pay the loan off PV 7000
in 8 years.) Whilst these programs can be useful make sure that if PMT –1283.584243
the course requires you to be able to use spreadsheets or recursive
FV 0
formulae you can do so.
P/Y 1
The display shows that repayments (PMT ) of $1283.584 243,
i.e. $1283.58 to the nearest cent, made once per year (P/Y ), will C/Y 1
see a loan of $7000 reduced to a future value (FV ) of $0 in 8 years
when compound interest is charged at 9.4% per annum, compounded annually, i.e. 1 compounding
per year (C/Y ). In the display the repayment is shown as a negative – it is money being taken
from us and put into the account. The $7000 on the other hand is shown as positive – it is money being
given to us in the form of a loan.
(Paying $1283.58 involves rounding the exact payment down, hence the final payment will be
a few cents more than $1283.58.)
Exercise 3E
1 Use the display on the right to determine what goes in the
blank spaces A to I in the following: Compound Interest
Repayments of A , i.e. B to the nearest cent, N 24

made C times per year, will see a loan of D reduced I(%) 9
to a future value of E in F years when compound
PV 4000
interest is charged at G% per year, compounded H ,
that is, I compoundings per year. Pmt –182.7389691
FV 0
PpY 12
CpY 12
2 At the beginning of one year Year Amount owed at Amount owed at

Jane borrows $10 000. She beginning of year end of year
repays $2000 at the end of that
1 $10 000 $10 000 × 1.1 - $2000 = $9000
year and at the end of each year
2 $9000 $9000 × 1.1 - $2000 = $7900
thereafter (except the last) until
the loan is paid off. Interest is 3 $7900 $7900 × 1.1 - $2000 = $6690
calculated at 10% per annum, 4 $6690 $6690 × 1.1 - $2000 = $A
compounded annually. The 5 $A $A × 1.1 - $2000 = $B
table on the right shows 6 $B $B × 1.1 - $2000 = $2284.39
the progress of the loan
7 $2284.39 $2284.39 × 1.1 - $2000 = $C
(to the nearest cent).
8 $C $C × 1.1 - $D = $0
Determine A, B, C and D.

3 Ranii borrows $1600 at the beginning of one month and repays $300 at the end of that month,
and every month thereafter (except the last) until the loan is repaid.
Interest is calculated at 1.5% per month.
The table below shows the progress of the loan (to the nearest cent).
Month Amount owed at Amount owed at

beginning of month end of month
1 $1600 $1600 × A - $300 = $1324
2 $1324 $1324 × A - $300 = $B
3 $B $B × A - $300 = $759.52
4 $759.52 $759.52 × A - $300 = $470.91
5 $470.91 $470.91 × A - $300 = $177.97
6 $177.97 $177.97 × A - $C = $0
Determine the values of A, B and C.
Use recursion techniques for questions 4, 5 and 6.
4 If $A is borrowed at a monthly interest rate of 1%, with $P

being repaid at the end of each month, the amount owing after
(n + 1) months is related to the amount owing after n months
ysedih/moc.kcotsrettuhS
according to the rule
Tn 1 = 1.01(T n) - P,
+
for n ≥ 0,
where T0 = A,
and $Tn is the amount owing after n months.
Find the amount owing at the end of each of the first four months if $50 000 is borrowed at
a monthly interest rate of 1% and $800 is repaid at the end of each month.
5 Hisham takes out a loan for $24 000 agreeing to repay $350 at the end of each monthly period,
after monthly interest is added at 1% of the outstanding balance. How much will he still owe on
this loan after 5 years?
Suppose instead that Hisham had made monthly repayments of $375. How much would he still
owe after 5 years in this case?
6 Jackson needs to borrow some money from a finance company

so that he can purchase a car costing $11 600. He is considering
oidutSkelAV/moc.kcotsrettuhS
three different repayment schemes, all of which involve him in
making regular monthly repayments and with interest added
at the rate of 15% per year (i.e. 1.25% per month) calculated
monthly on the outstanding balance before the repayment
is made.
Scheme A: Borrow the full $11 600 and make monthly repayments of $250.
Scheme B: Borrow the full $11 600 and make monthly repayments of $290.
Scheme C: Pay $1200 that he has saved towards the cost of the car,
borrow the rest and make monthly repayments of $290.
How long will each scheme take to pay off the loan?

Use the inbuilt financial programs available on some calculators or a suitable online financial
calculator to answer each of questions 7, 8 and 9.
7 To the nearest cent what constant monthly amount needs to be repaid to see a loan of $10 000,
with compound interest of 6% per annum compounded monthly, paid off in 2 years?
10 The spreadsheet below left is for the situation of borrowing $4000, paying interest of 1% per
month and repaying $100 per month. The spreadsheet below right shows the same situation but
now with $150 repaid each month.
A B C D A B C D
1 Amount borrowed $4,000.00 1 Amount borrowed $4,000.00
2 Interest rate per month 1.00% 2 Interest rate per month 1.00%
3 Monthly repayments $100.00 3 Monthly repayments $150.00
4 Balance at end of month 1 $3,940.00 4 Balance at end of month 1 $3,890.00
5 2 $3,879.40 5 2 $3,778.90
6 3 $3,818.19 6 3 $3,666.69
7 4 $3,756.38 7 4 $3,553.36
8 5 $3,693.94 8 5 $3,438.89
9 6 $3,630.88 9 6 $3,323.28
10 7 $3,567.19 10 7 $3,206.51
11 8 $3,502.86 11 8 $3,088.58
12 9 $3,437.89 12 9 $2,969.46
13 10 $3,372.27 13 10 $2,849.16
14 11 $3,305.99 14 11 $2,727.65
15 12 $3,239.05 15 12 $2,604.92
16 13 $3,171.44 16 13 $2,480.97
17 14 $3,103.15 17 14 $2,355.78
18 15 $3,034.19 18 15 $2,229.34
19 16 $2,964.53 19 16 $2,101.63
20 17 $2,894.17 20 17 $1,972.65
21 18 $2,823.12 21 18 $1,842.38
22 19 $2,751.35 22 19 $1,710.80
23 20 $2,678.86 23 20 $1,577.91
24 21 $2,605.65 24 21 $1,443.69
25 22 $2,531.70 25 22 $1,308.13
26 23 $2,457.02 26 23 $1,171.21
27 24 $2,381.59 27 24 $1,032.92
Create such a spreadsheet yourself, or adapt one you have already created, and find how much
needs to repaid per month in order to pay the loan off in exactly two years.
Suppose instead the interest rate was 1.2% per month. How much would need to be repaid per
month now to repay the loan in exactly two years?

What price house can these people afford?
Let us suppose that in order to purchase a house costing $480 000 John and Maxine pay a deposit
of $30 000 from their savings and borrow the remaining $450 000 from a financial institution.
The institution charges a fixed 8% per annum compound interest with compounding occurring
monthly. John and Maxine arrange to repay the loan by making regular monthly payments for
twenty-five years.
grebmoolB/segamI ytteG
Using a calculator with the ability to perform financial calculations,
Compound Interest
or a similar financial capability on a computer or online, we can
determine that their monthly repayments would be $3473.17, to N 300
the nearest cent. I(%) 8
Before entering into such an arrangement John and Maxine PV 450000
would probably have first worked out what they could afford
to pay per month. PMT –3473.172987
Let us suppose that Kym and Hasim have $35 000 saved to put FV 0
down as a deposit on a house. They feel they can afford to pay P/Y 12
$2000 per month towards the cost of a loan. They plan to get
C/Y 12
a loan for a term of 25 years making regular monthly payments
and are told the interest rate is 7.6% per annum compounded
monthly.
What value house can they afford?
Compound Interest
As the display on the right indicates, Kym and Hasim can afford
N 300
a loan of $268 000, to the nearest $1000.
I(%) 7.6
Therefore, with their deposit of $35 000, they could afford a house
costing $303 000 (= $268 000 + $35 000). PV 268273.4425
PMT –2000
FV 0
P/Y 12
C/Y 12

In the case of Kym and Hasim if they borrow $268 000 they will
make 300 repayments each of $1997.96 (see the display on the right). Compound Interest
Hence their total repayments will be N 300
300 × $1997.96 = $599 388. I(%) 7.6
Thus on their loan of $268 000 they have paid PV 268000

$599 388 - $268 000 = $331 388 interest, PMT –1997.961464
… or have they? Read on. FV 0
By switching to ‘Amortization’, as shown below right, some P/Y 12
calculators show this total interest amount, and other information
about the loan. Some calculators, and some online amortization C/Y 12
calculators, can display a table showing month-by-month details
about the loan.
Amortization
Wanting the total interest paid from payment 1 to payment 300
PM1 1
the display on the right shows PM1 as 1 and PM2 as 300 and
displays ∑INT, as shown. PM2 300
If we round the ∑INT amount shown in the display to the nearest I% 7.6
dollar, we again get $331 388. PV 268000
Wondering where the extra $0.4392 came from?
PMT –1997.961464
Well, one calculation used 300 lots of $1997.96 and the other used
P/Y 12
300 lots of $1997.961 464 ….
Explore your calculator with regard to amortization and investigate C/Y 12
online amortization calculators. BAL
• The slight rounding down of $1997.961 464 to $1997.96 means INT
that in theory there would still be a very small amount ($1.3051)
PRN
remaining on the loan after the 300th payment. In practice, to
clear the loan on the 300th payment, the final payment would be ΣINT –331388.4392
$1997.96 + $1.31, i.e. $1999.27. ΣPRN
The total interest amount of $331 388
(= 300 × $1997.96 - $268 000) quoted at the top of the page has not allowed for this change in the
final payment. Thus a more accurate calculation of the total interest would be
$331 389.31 (= 299 × $1997.96 + 1 × $1999.27 - $268 000),
a total directly obtainable from ∑INT in the amortization calculation with PM1 = 1, PM2 = 300
and monthly repayment of $1997.96.
However, the $331 388 would often be considered a satisfactory approximation.
• Under different rounding regimes slight variations are possible.
If we round each month to the nearest cent, as some online If you do ever think of taking out
amortization calculators do, the final payment becomes a loan to buy a house you should
$1999.06 and the total interest is then seek expert advice specific to
your situation. The work here just
$331 389.10 (= 299 × $1997.96 + 1 × $1999.06 - $268 000). introduces the mathematics.
• Amortization can also be written amortisation. However,
we will use the z spelling here as that is how your calculator
is likely to show it.

Exercise 3F
What price house can each of the following people afford? In each case assume that the interest rate
shown is per annum, compounded monthly.
Amount they can afford Deposit Duration Interest

to repay per month saved of loan rate
1 Fran and Michael $3800 $17 000 25 years 7.5%
2 Peta and Peter $2400 $20 000 25 years 8.2%
3 Rania and Umar $4000 $35 000 20 years 9.1%
4 Hue and James $2000 $450 000 15 years 6.2%
5 Kirra $2650 $30 000 30 years 7.9%
6 Juan and Denise borrow $450 000 at a fixed interest rate of 8.34% per annum, compounded
monthly. They plan to pay off the loan in exactly 25 years by making the same monthly
repayments for 25 years.
How much will each monthly repayment be (nearest cent) and how much interest will they pay
in total (nearest $10)?
7 Chris and Terri borrow $380 000 at a fixed interest rate of 7.2% per annum, compounded monthly.
They plan to repay the loan by making the same regular repayments each month for twenty years.
How much will each repayment be (nearest cent) and how much will they repay in total
(i.e. interest plus loan repayment) over the twenty years, to the nearest $10?
How much quicker would they have paid off the loan if they had paid $100 per month more?
INVESTIGATE
In the above cases we have only considered the cost of the house. However other expenses are
incurred when purchasing a house, for example, stamp duty, insurance, settlement agent fees,
purchaser’s share of the rates. Investigate and write a brief report.
Miscellaneous exercise three

work of any previous chapters, and the ideas mentioned in the Preliminary work section at
the beginning of the book.
1 $8000 is invested for 5 years with an annual compound interest rate of 8%. Find the amount this
account is worth at the end of the 5 years if the compounding occurs
a annually, b quarterly, c monthly.
2 What does it mean if the seasonal index for Autumn is 1.08 (or 108%)?

3 Deseasonalise the following raw data given the seasonal indices are as stated.
Give the deseasonalised data to the nearest integer.
Season Mon Tue Wed Thur Fri Sat

Raw data 113 79 84 151 151 160
Seasonal index 0.92 0.70 0.68 1.21 1.29 1.20
4 Find the first five terms of each of the following recursively defined sequences.
a T1 = 7, Tn + 1Tn + 12.
= b T 1 = 100, Tn +
T n - 15.
1 =
c T1 = 5000, Tn +1 = 1.2 × Tn . d T 1 = 2000, T n = T n 1 ÷ 10.

-
e T1 = 4, Tn = 2Tn 1 + 3. -
f T 1 = 200, T n 1 = 1.5Tn - 4.
+
5 Which of the following graphs show

a an outlier? b seasonal variation? c an increasing trend?
d a decreasing trend? e no irregular fluctuations?
Graph A Graph B
100 100
80 80
60 60
40 40
20 20
5 10 15 t 5 10 15 t
Graph C Graph D
100 100
80 80
60 60
40 40
20 20
5 10 15 t 5 10 15 t
6 Use recurrence relation techniques for this question.

Johan borrows $50 000 with interest charged at 7.5% per annum on the declining balance,
compounded monthly. At the end of the first period of one month, and at the end of every
period of one month thereafter, Johan repays $1000 on the loan.
How much will Johan still owe on this loan immediately after the interest has been added and
the $1000 has been credited for the end of
a month two? b month six? c year one? d year two?

7 With the aid of a calculator with a built-in facility to carry out compound interest calculations,
or otherwise, determine each of the following.
a The value at the end of five years of an initial investment of $12 000 invested at 6.4% per
annum compound interest, compounded quarterly.
b The number of years required for an initial investment of $25 000 invested at 7.5% per
annum compound interest, compounded annually, to be worth $75 000.
c The number of years required to pay off a loan of $400 000 if interest of 8% per annum is
compounded monthly and $3500 is paid off the loan at the end of every month.
d The annual (nominal) compound interest rate required to see an initial investment of $25 000
exceed $125 000 within fifteen years if interest is compounded
i annually,
ii monthly.
8 In 1971 one Australian dollar, A$, could purchase approximately 400 Japanese yen.
In 2012 one Australian dollar could purchase approximately 80 Japanese yen.
The average numbers of yen, ¥, that could be purchased for each A$ 1 in each of the years from
1971 to 2012 were approximately as follows:
Year 1971 1972 1973 1974 1975 1976

A$ 1 would buy ¥ 400 ¥ 372 ¥ 367 ¥ 408 ¥ 405 ¥ 379
Year 1977 1978 1979 1980 1981 1982

A$ 1 would buy ¥ 331 ¥ 274 ¥ 227 ¥ 260 ¥ 248 ¥ 257
Year 1983 1984 1985 1986 1987 1988

A$ 1 would buy ¥ 233 ¥ 211 ¥ 193 ¥ 141 ¥ 101 ¥ 100
Year 1989 1990 1991 1992 1993 1994

A$ 1 would buy ¥ 107 ¥ 112 ¥ 107 ¥ 100 ¥ 83 ¥ 73
Year 1995 1996 1997 1998 1999 2000

A$ 1 would buy ¥ 70 ¥ 77 ¥ 90 ¥ 86 ¥ 78 ¥ 68
Year 2001 2002 2003 2004 2005 2006

A$ 1 would buy ¥ 61 ¥ 66 ¥ 70 ¥ 79 ¥ 80 ¥ 85
Year 2007 2008 2009 2010 2011 2012

A$ 1 would buy ¥ 93 ¥ 99 ¥ 75 ¥ 81 ¥ 82 ¥ 81
(ABARES 2013, Agricultural commodity statistics 2013. CC BY 3.0)
With the assistance of a computer spreadsheet, display the above data graphically, together with
the five-point moving averages.
Write a few sentences summarising the data.
How does today’s exchange rate compare?

4.
Finance II:
Drawing down
the investment
• Superannuation
• Annuities
• Indexing
• Perpetuities
• Suppose the frequency of payments
≠ the frequency of compounding
• Miscellaneous exercise four

Superannuation
Retirees in Australia may, if they satisfy certain conditions, qualify for the government age pension.
The amount they receive depends on what other income and assets they have. However, the total of
these pension payments form a considerable expenditure for the government and, with increasing
life expectancy, an expenditure that could get bigger and bigger in years to come. For this reason
the government encourages people to put money into savings during their years of employment so
that they will have their own money to support themselves in retirement, and hence only need a part
pension or perhaps even no pension payment at all from the government. Saving to have money
available in retirement is called superannuation. During a person’s working years they accumulate
funds in their superannuation account and then, when they retire from work, they draw down those
funds as an income they can live on, supplemented by the government pension if the requirements of
an assets and income test are met.
segamI ssenisuB yeknoM/moc.kcotsrettuhS

87zemis/moc.kcotsrettuhS
INVESTIGATE
How much does a person need to have saved in superannuation to live comfortably in retirement?
What sort of annual income will a retired couple require?
What does the internet suggest as answers to these questions?
When a person is employed their employer must pay a certain percentage of their wage into
a superannuation account. Thus during the employment years a person’s superannuation account is
in its accumulation phase. Then, when the employee retires, they draw money from the superannuation
account as a self-funded pension, possibly supplemented by the government pension, to give them an
income in retirement. The superannuation account is then in its pension phase. Of course in both the
accumulation phase and the pension phase any funds in the account can earn interest. Thus in the
pension phase interest is earned by the account and, at the same time, money is periodically withdrawn
from the account.
ISBN 9780170395069 4. Finance II: Drawing down the investment 77

WS EXAMPLE 1
Annuities At retirement Susie invests the $750 000 she has saved in her superannuation account in an account
giving interest of 6% per annum compounded annually from which she will withdraw $60 000 after
one year, and after each year thereafter. How much will be left in the account immediately after the
tenth withdrawal (rounded to the nearest dollar)?
If she wants the account to last her at least 35 years, withdrawing a constant amount each year,
what should this constant amount be (rounded down to a multiple of $100)?
Solution
Initial amount invested: $750 000
Account balance after 1 year: $750 000 × 1.06 - $60 000
Account balance after 2 years: ($750 000 × 1.06 - $60 000) × 1.06 - $60 000
We have Tn + 1 = Tn × 1.06 - $60 000,
and T0 = $750 000
Using the recursion capability of some calculators,
a spreadsheet or the inbuilt financial programs of some an+1 = 1.06•an – 60000
calculators, we find that: a0 = 750000
Immediately after the tenth withdrawal there will be bn+1 =
$552 288 in the account, to the nearest dollar. b0 = 0
cn+1 =
c0 = 0
n an
6 6.5E+5
Compound Interest 7 6.2E+5
8 6.0E+5
N 35 9 5.8E+5
10 5.5E+5
I(%) 6
552288.075864291
PV –750000
Pmt 51730.39423
FV 0 Varying the withdrawal amount to reduce the balance to zero in
35 years, or using a calculator with an inbuilt financial capability,
PpY 1
we can determine that Susie can withdraw up to $51 700 each
CpY 1 year (rounded down to a multiple of $100) for the account to
last at least 35 years.
rimogarD nailuI/moc.kcotsrettuhS

The graph below shows the declining balance in the account of the previous example which reduced
the balance to zero in 35 years by withdrawing $51 730.39 per year.
$750 000 Initial value of $750000

$700 000
$650 000
$600 000
$550 000
tnuocca fo ecnalaB
$500 000
$450 000
$400 000
$350 000
$300 000
$250 000
$200 000
$150 000
$100 000
$50 000 Annual payment of $51 730.39
5 10 15 20 25 30 35
Can your calculator display a graph of this type from the terms of a recursively defined sequence? Investigate.
Annuities
In the previous example Susie managed her own income stream by making a deposit into an account,
withdrawing a fixed amount each year to live on and allowing the balance of the account to earn
interest. A compound interest investment from which regular payments are made for a fixed period
of time is called an annuity.
Making regular payments to pay off a loan, as encountered in the previous chapter, is a bit like an annuity.
The finance company invests money with us in the form of a loan and we make regular payments back.
However the word annuity is usually used in the context of using a sum of money to set up a regular income.
By making a one-off payment a person can enter into an annuity arrangement whereby they will
receive a guaranteed fixed regular amount for an agreed period of time. Such arrangements can be very
useful for generating a steady income in retirement.
Fixed term annuities make regular payments for a fixed period of time. If the person who purchased
the annuity dies before the end of the fixed period of time the remaining payments, or the remaining
balance, would pass to a beneficiary.
Whole life annuities make regular payments to the purchaser of the annuity for as long as they live.
Under this sort of annuity the payments stop when the person dies, i.e. upon the death of the purchaser
such an annuity has no residual value. If the person dies soon after the annuity commences the finance
company ‘wins’ but if the person lives well beyond normal life expectancy the person ‘wins’. Life
annuities can be seen as an insurance in case one lives a long time. Even if the person does live a long
time, regular income is assured. The finance company spreads the risk of paying out for more years
than expected by having annuities with a lot of people.

Indexing
In some cases the regular amount paid out from an annuity might increase over time so that the
amount received keeps up with inflation and maintains its ‘buying power’. We say that the regular
payments are indexed, i.e. the payments change in line with the movement of some price index, e.g. the
consumer price index, a measure of the increase in the price of a typical household selection of goods.
EXAMPLE 2
A person wishes to deposit $200 000 into an account that will earn interest of 8% per annum,
compounded annually, withdrawing $25 000 at the end of the first year, $25 500 at the end of the
second year, $26 010 at the end of the third year, and so on with each annual withdrawal being
a 2% increase on the withdrawal of the previous year.
a What will be the balance of the account immediately after the fifth withdrawal?
b How long will it take for the balance in the account to reduce to zero?
Solution
a Initial amount invested: $200 000
Account balance after 1 year: $200 000 × 1.08 - $25 000 I.e. $191 000
Account balance after 2 years: $191 000 × 1.08 - $25 000 × 1.02 I.e. $180 780
Account balance after 3 years: $180 780 × 1.08 - $25 000 × 1.022 I.e. $169 232.40
We have Tn + 1 = T n × 1.08 - $25 000 × 1.02n,
and T0 = $200 000
Using a calculator able to display the terms of a recursively defined sequence, or a spreadsheet,
or by working the terms through:
T4, the account balance after 4 years = $156 240.79 (nearest cent)
Immediately after the fifth withdrawal the balance will be $141 679.25.
b Continuing the sequence:

After a further year this balance will grow to $12 884.18 × 1.08
= $13 914.91
The balance will reduce to zero at the end of the 12th year when the withdrawal will be just
$13 914.91.

Note: Care needs to be taken when using online annuity calculators. Some annuities work on the
basis that the first payout from the annuity occurs when the annuity is opened, e.g. as soon as
a superannuation amount is rolled into pension phase, and some work on the basis that the first
payout occurs at the end of the first month or year, whatever the agreed frequency of payment is.
Perpetuities
Suppose we invest $5000 in an account paying 8% per annum compounded annually, and we withdraw
$400 at the end of each year.
Initial amount invested: $5000
Account balance after 1 year: $5000 × 1.08 - $400 I.e. $5000.
Account balance after 2 years: $5000 × 1.08 - $400 I.e. $5000.
Account balance after 3 years: $5000 × 1.08 - $400 I.e. $5000. Etc.
In this case the annual withdrawal of $400 could continue forever because each year the interest earned
exactly equals the amount paid out. This sort of annuity is called a perpetuity, the word perpetuity
meaning lasting forever.
Consider the case, for example, of someone wishing to fund an annual award of $400 to be given to
a deserving student in a school for some reason. By donating $5000 into an account paying 8% per
annum the interest produced will fund the award forever, i.e. in perpetuity, endlessly.
Exercise 4A
1 An ex-pupil of a school wishes to invest a sum of money into an account paying 7.5% interest,
compounded annually, so that every year thereafter $600 can be made available from the account
as an award to a deserving student.
Explaining your reasoning determine how much money needs to be invested to make this
perpetual award possible.
2 Someone who has made their fortune from mining wishes to set up an account that will allow
a research grant of $75 000 to be paid every year thereafter to fund research at a University School
of Mines Department. So that the grant can be perpetual the account, which pays 8% per annum
interest, compounded quarterly, needs to generate annual interest of at least $75 000.
Explaining your reasoning determine how much the initial ‘one-off’ investment needs to be
(rounded up to the next dollar).
3 What initial investment is required into an account that pays annual interest of 6%, compounded
monthly, if the interest earned each year is to pay for an annual perpetual award of $15 000?

4 Let us suppose that $P is invested in an account that pays R% per annum, compounded annually.
At the end of the first year, and every year thereafter, $A is withdrawn from the account and after
exactly Y years the balance of the account reaches zero.
For each of the following, state the option that will increase Y , other quantities remaining the same.
a Increase P or decrease P.
b An increase in R or a decrease in R.
c An increase in A or a decrease in A .
5 At retirement, Kelvin invests the $620 000 he has in his superannuation account in an account
giving interest of 5.8% per annum compounded annually, from which he will withdraw $50 000 at
the end of every twelve months to live on. Using a recursive rule determine how much will be left
in the account immediately after the tenth withdrawal (rounded to the nearest ten dollars).
For how many years will Kelvin be able to withdraw $50 000 per year?
For how many years would Kelvin have been able to withdraw $50 000 per year if instead the
interest rate had been 7.8%?
6 Julie is left $765 000 in a will and decides to invest $700 000 of this money in an account paying
5.4% interest compounded annually. Julie plans to retire early and use this money to live on by
withdrawing $50 000 at the end of the first year and each year thereafter until the account balance
is exhausted. Use a recursive rule to determine for how many years will she be able to withdraw
$50 000 from this account.
If instead Julie withdrew $45 000 at the end of each year for how many years could she do this?
Can you foresee any problems with the plan to withdraw a constant amount each year for a long
period of time?
7 Use the financial capability of some calculators or computer programs to determine, to the nearest
dollar, how much should be invested in an annuity earning 12% annual interest, compounded
annually, to provide a regular annual income of $45 000 for exactly 25 years. (The first payment is
to be at the end of the first 12 months).
8 Repeat the previous question but now for an interest rate of just 6% per annum compounded
annually.
9 Use the financial capability of some calculators or computer programs to determine, to the nearest
dollar, how much should be invested in an annuity earning 8% annual interest, compounded
monthly, to provide a regular monthly income of $3000 for exactly 20 years. (The first payment is
to be at the end of the first month).
10 Repeat the previous question but now for an interest rate of just 5% per annum compounded
annually.

11 A finance company offers an annuity in which a person deposits a lump sum of $300 000 into an
account and interest is compounded annually at the rate of 7.5% per year. At the end of the first
year, and every year thereafter for as long as the balance in the account allows, the annuity pays out
$30 000 with the final payment, which might be less than $30 000, reducing the balance to zero.
(In the event of death the remaining funds will be paid into the estate of the deceased.)
a Use a recurrence rule to determine how many $30 000 payments will be made and what the
final payment will be (assuming the person does not die in the meantime).
b Repeat part a but now determine the answer using a calculator with financial capability.
12 A finance company offers an annuity in which a person deposits a lump sum of $500 000 into
an account and interest is compounded monthly at the rate of 0.4% per month. At the end of the
first month and every month thereafter, the account pays out $4000 with the final payment, which
might be less than $4000, reducing the balance to zero. (In the event of the death of the annuity
holder remaining funds will be paid into the estate of the deceased.)
a Use a recurrence rule to determine how many $4000 payments will be made and what the
final payment will be (assuming the person does not die in the meantime).
b Repeat part a but now determine the answer using a calculator with financial capability
or computer based financial software.
13 To the nearest dollar, how much should be invested in an annuity paying 10% annual interest,
compounded annually, to provide a regular annual income of $42 000 paid at the end of each year,
for exactly 25 years?
Determine the answer by
a using a recursive formula with an initial ‘guess’ of $400 000
b using the financial capability of some calculators or computer programs.
14 Tony wins $500 000 on the lottery and decides to use $400 000 of it to purchase a 15-year annuity
that pays a fixed amount every year for fifteen years, commencing one year after the initial
purchase. The annuity adds interest to the account balance at 8% per annum, compounded
annually, and deducts the regular payment at the end of each year. At the end of the 15 years
the account balance is zero.
How much does Tony receive each year (to the nearest dollar)?
a using a recursive formula with an initial ‘guess’ of $50 000
jalaT/moc.kcotSi

15 A person wishes to deposit $300 000 into an account that will earn interest of 8% per annum,
a What will be the balance of the account immediately after the third withdrawal?
16 A person wishes to deposit $250 000 into an account that will earn interest of 7.5% per annum,
a What will be the balance of the account immediately after the eighth withdrawal?
Suppose the frequency of payments ≠ the

frequency of compounding
So far in this chapter, in all of the annuity situations (apart from two questions involving perpetuities)
the frequency of the payments, e.g. annual, quarterly, monthly, has been the same as the frequency of
the compounding periods, e.g. annual, quarterly, monthly. Let us now consider situations in which
compounding occurs more frequently than the payments are made (see example 3) and situations in
which payments occur more frequently than compounding (see example 4)
EXAMPLE 3
Suppose that $250 000 is invested into an account paying 8% interest, compounded quarterly,
with $40 000 withdrawn from the account at the end of the first year and at the end of every
year thereafter.
Determine the balance of the account immediately after the sixth withdrawal.
Solution
Initial amount invested: $250 000
Account balance after 1 year: $250 000 × 1.024 - $40 000 I.e. $230 608.04.
Account balance after 2 years: $230 608.04 × 1.024 - $40 000 I.e. $209 617.56 (nearest cent)
We have Tn + 1 = Tn × 1.024 - $40 000,
and T0 = $250 000

Displaying the terms of this sequence on a calculator,
as shown on the right, we see that the balance of the account an+1 = 1.024•an – 40000
immediately after the sixth withdrawal is $106 866.65 a0 = 250000
(to the nearest cent).
bn+1 =
Alternatively the same answer could be obtained using b0 = 0
the inbuilt financial capability of some calculators: see cn+1 =
the typical display below. c0 = 0
n an
Compound Interest
2 2.1E+5
N 6
3 1.9E+5
I(%) 8 4 1.6E+5
5 1.4E+5
PV –250000 6 1.1E+5
PMT 40000 106866.654903457
FV 106866·6549
P/Y 1
C/Y 4
Notice that in the previous example our use of the multiplication factor 1.024 brought our quarterly
compounding up to its equivalent effective annual compounding and thereby had the frequency of
compounding matching the frequency of payment.
Now suppose that payments occur more frequently than compounding. For example payments
occurring every quarter, compounding occurring annually and an annual interest rate of 8%. We
can again adjust the compounding period to match the payment period by finding the quarterly
multiplication factor that will produce an annual compounding of 8%. This quarterly multiplication
factor is 1.019 426 547 because,
if x4 = 1.08 then x = 1.019 426 547.
However, whilst we would need this equivalent multiplication factor if we were using a recursive
approach, the use of a calculator with built-in financial capability gives a ready alternative that avoids us
having to calculate this figure, as shown in the next example.
rimogarD nailuI/moc.kcotsrettuhS

EXAMPLE 4
Suppose that $250 000 is invested into an account paying 8% interest, compounded annually,
with $10 000 withdrawn from the account three months later and a further $10 000 each
quarter thereafter.
Determine the balance of the account after three years.
Solution
The display shows that after 3 years (12 quarterly
Compound Interest
payments of $10 000, i.e. 4 payments per year) an account
paying 8% interest per annum, compounded annually N 12
(1 compounding per year), and initially worth $250 000, I(%) 8
will be worth $181 238.775 after three years.
PV –250000
PMT 10000
FV 181238·775
P/Y 4
C/Y 1
Notice that using the 1.019 426 547 multiplication factor

an+1 = 1.019426547•an – 10000
mentioned on the previous page to create an appropriate
a0 = 250000
recursive rule gives this same answer, as shown below.
However, for these questions, in which the payments n an
occur more frequently than the compounding, using the 8 2.1E+5
financial capability of some calculators is recommended 9 2.0E+5
rather than using a recurrence relation. 10 1.9E+5
11 1.9E+5
12 1.8E+5
181238.775301663
87zemis/moc.kcotsrettuhS

Exercise 4B
1 Use a recursive approach for this question.
Suppose that $400 000 is invested into an account paying 6% per annum interest, compounded
semi-annually, with $35 000 withdrawn from the account at the end of the first year and at the
end of every year thereafter.
Determine the balance of the account immediately after the tenth withdrawal.
2 For this question use a recursive approach (or a spreadsheet of your own creation) and make
$500 000 your ‘first guess’ at the amount invested.
To the nearest dollar, how much needs to be invested in an account paying interest of 6% per
annum, compounded quarterly, for $45 000 to be withdrawn at the end of every year for 25 years?
3 For this question use the financial capability of some calculators.

How much needs to be invested, to the nearest cent, in an account paying interest of 9% per annum,
compounded annually, for $10 000 to be paid at the end of the first 3 months, and every three
months thereafter for 15 years, at the end of which time the account has zero balance remaining?
4 For this question use the financial capability of some calculators.

How much needs to be invested, to the nearest cent, in an account paying interest of 6% per
annum, compounded annually, for $4000 to be paid at the end of the first month, and every
month thereafter for 20 years, at the end of which time the account has zero balance remaining?
5 Upon retirement John switches his superannuation fund of $480 000 into an account paying
9% per annum compounded monthly. After three months, and every three months thereafter John
wants the account to pay him $15 000, for as long as possible.
How many payments of $15 000 will this scheme allow John to receive and what will be the final
payment that closes the account?
RESEARCH
Do some research and write a brief report on some of the following questions.
• What incentives and rules does the Government of Australia put in place to ensure / encourage
people to put money into superannuation?
• What is the current average balance that people have in superannuation?
• Why is an ageing population a challenge for governments?
• What is the retirement age in Australia? What is it in other countries?

Computer spreadsheet or graphic calculator activity
Sue wins $120 000 in a lottery. Rather than spend it all
immediately she decides to use some of it, and perhaps
even all of it, to allow herself to take some years off work,
rebmoCcMsalociN/moc.kcotSi
or at least reduce her hours somewhat.
She asks for your advice regarding how long the money would
last under each of five schemes she has in mind.
Determine how many months she could withdraw the stated
monthly withdrawal in each of the following schemes.
Scheme 1 Put the money in a box under the bed and withdraw $1500 each month to live on.
Scheme 2 Invest the entire $120 000 in an account paying 8% interest per year, compounded
monthly, and withdraw $1500 at the end of each month to live on.
Scheme 3 Same as scheme 2 but this time suppose the account pays interest of 9% per annum
compounded monthly.
Scheme 4 Use $20 000 of the $120 000 to pay for flights
and accommodation for an around the world holiday,
invest the rest in an account earning 9% per annum
otohPnakuS/moc.kcotsrettuhS
interest, compounded monthly, and withdraw $1500
each month during the holiday and thereafter.
Scheme 5 Invest the entire $120 000 in

an account paying 9% interest
per year, compounded monthly.
At the end of each of the first
12 months withdraw $1500,
at the end of each of the next
12 months withdraw $1600,
at the end of each of the next
12 months withdraw $1700
and so on, increasing the
monthly withdrawal by $100
after each 12-month period.
lsxela/moc.kcotSi

Miscellaneous exercise four
This miscellaneous exercise may include questions involving the work of this chapter,
the work of any previous chapters, and the ideas mentioned in the Preliminary work section
at the beginning of the book.
1 How much interest is earned in four years if $4000 is invested in an account paying 12% per
annum compounded monthly?
2 A time series graph showing the number of people attending a particular music festival each year,
for ten years, suggested that linear regression could be used to summarise the relationship between
N, the numbers attending, and t the year number where t = 1 is used for the first year in the ten-
year period.
The least squares regression line for the data had equation:
N = 855t + 9459
Interpret the ‘855’ in this equation.
3 Giving each answer as a percentage correct to three decimal places, calculate the effective interest
rate if the annual nominal rate is 6% and interest is compounded
a quarterly, b monthly, c daily.
4 To the nearest cent, what ‘one-off’ lump-sum payment must be invested into an account paying
compound interest of 8% per annum compounded quarterly for the investment to be worth $5000
two years later?
5 If we use the recursion relation Tn 1 = Tn × 1.065, T0 = $4300, the value of Tn gives the value of an
+
account n years after a deposit was made to open the account, with interest compounded annually.
a What was the amount of the deposit?
b What was the annual interest rate?
c Find the balance of the account after interest had been added at the end of the sixth year
(to the nearest cent)?
d What annual interest rate would achieve the balance that was the answer to part c but after
just two years? (Give your answer as a percentage correct to three decimal places).
6 The table below relates to a savings account which Waylon opens by making an initial deposit and
then adding a further fixed amount at the end of each month.
Interest earned is compounded monthly with the same monthly interest rate throughout.
Month Amount in account at Amount in account at end of month,

beginning of month following adding of interest and deposit
1 $1600 $1874.00
2 $1874.00 $2152.11
Determine the initial investment, the fixed monthly interest rate and the fixed monthly deposit.

7 a Determine the values that go in the spaces marked A, B, … G in this table.
t Year Months of Sales Four month Sales ÷ (four month

year mean for year mean for the year)
1 First 4 months 442 1.1190
2 2010 Second 4 months 299 395 A
3 Third 4 months 444 1.1241
5 2011 Second 4 months 241 300 B
8 2012 Second 4 months 203 261 C
10 First 4 months 213 0.9953
11 2013 Second 4 months 171 214 D
12 Third 4 months 258 1.2056
13 First 4 months 170 1.0625
14 2014 Second 4 months 122 E F
15 Third 4 months 188 G
b Create a completed version of the following tables.

Calculation of seasonal indices
2010 2011 2012 2013 2014 Seasonal index
(3 decimal places)
1st 4 months 1.1190 1.0233 1.1264 0.9953 1.0625 1.065
2nd 4 months
3rd 4 months 1.1241 1.1733 1.0958 1.2056
Seasonally adjusted sales figures, S (rounded to the nearest integer)

2010 2011 2012 2013 2014
1st 4 months
2nd 4 months
3rd 4 months
c Use the ordered pairs (t, S) to determine the equation of the regression line
S = At + B.
d Use this regression equation and the seasonal indices to predict sales for the three 4-month
intervals of 2015.

5.
Minimum
spanning trees
• Spanning trees
• Minimum spanning trees – a systematic approach
• Minimum spanning tree from the distances table
• Miscellaneous exercise five
Spanning trees
The network on the right shows the B 22 E 10 H
roads that exist between nine farms,
A to I, with the numbers indicating 11 7 10 9 10
8
distances in kilometres. The water
16 D 12
authority wish to connect these farms A
12 G 7
I
C 7
to the mains water supply by laying 21 8
water pipes alongside existing roads. F
This is not the same as finding the B 22 E 10 H

shortest route from A to I because
this shortest route, shown on the 11 10 9 10
7 8
right, does not connect up all the
16 D 12
farms (B, E, F and H are not ‘on line’). A
12 7
I
C 7 G
21 8
F
Instead we need to find a system that B 22 E 10 H

includes all the farms but which has
no unnecessary connections. What we 11 10 9 10
7 8
need is a minimum spanning tree for
16 D 12
this network. A
12 7
I
C 7 G
21 8
F
(A spanning tree of a graph is a subgraph that connects all of the vertices and is itself a tree.)
The spanning tree shown requires a total length of 85 km (= 11 km + 7 km + 22 km + 8 km + 10 km
+ 9 km + 10 km + 8 km).
Can you find a shorter spanning tree?

Can you find the shortest spanning tree?
How long is it?
If you think you have found the minimum spanning tree for the network shown above, then try to find
it for the network shown below.
E
23
B 35
12
25
23 30
10 H
25 F
20 C
A 17
34 27
15
33
21 G
D
ISBN 9780170395069 5. Minimum spanning trees 93

WS
Minimum spanning trees – a systematic approach
Consider the network on the right. B 40 C
Minimum
spanning trees
To determine the minimum spanning tree 38 18 52 30
systematically we can in fact start at any vertex but we 22
D
25 G
will start at A as that seems a logical place to start. A 36 H
20 35 24 20
From A we next bring ‘on line’ the vertex that is A’s 34
E
nearest neighbour. Point G, being just 25 units from F 75
A is the nearest neighbour in this example.

Now that we have A and G on line we next connect to B 40 C
the vertex that is closest to one of these two vertices. 30
38 18 52
In our example, B is the next nearest neighbour to one 22
of our on line points A and G. 25 G D
A 36 H
20 35 24 20
34
E
F 75
Now that we have A, G and B on line we now look B 40 C

for the vertex that is not yet on line but that is the 38 30
18 52
nearest neighbour to one of our on line points. 22
25 G D
Thus point F, being just 20 units from G, is the A 36 H
20 24 20
next nearest neighbour. 35
34
E
F 75
We now have A, G, B and F on line and again look B 40 C

for the next nearest neighbour. In this case point A is 38 30
18 52
only 34 units from F but we already have A on line so
22
this connection would be pointless. (It would give an 25 G D
A 36 H
unwanted cycle in our connection.) Thus we choose to 20 35 24 20
34
bring H on line as it is the nearest neighbour not yet E
F 75
on line.
Continuing in this way until all points are on line gives B 40 C
the minimum spanning tree shown on the right. 30
38 18 52
22
25 G D
Of course this process would usually be carried out A 36 H
on one copy of the network. The five copies used 20 35 24 20
34
here are purely to illustrate the steps of the method. E
F 75
The reader should now use this method on a copy B 40 C

of the network shown on the right but this time start 30
38 18 52
at a vertex other than A and confirm that a different
22
starting point still leads to the same minimum 25 G D
A 36 H
spanning tree. 20 35 24 20
34
E
F 75

The method explained on the previous page, in which we:
choose any vertex as our initial ‘on line’ vertex and then build up the spanning tree by connecting online
vertices to the ‘nearest neighbour’, whilst always making sure that no cycles are introduced,
is called Prim’s algorithm.
An alternative approach is to use Kruskal’s algorithm. In this method we:
start with the ‘shortest’ edge, then choose the next’ shortest’ edge, even if it is not connected to the previously
selected edge, and then continue this process, always making sure that no cycles are introduced. When all
vertices are ‘on line’ we have the minimum spanning tree.
Using this Kruskal’s approach for the network on the previous page:
First we choose the ‘shortest’ edge in the network, B 40 C
in this case GB which has a weighting of just 18. 38 30
18 52
22
25 G D
A 36 H
20 24 20
35
34
E
F 75
Then we choose the next ‘shortest’ edge. In this case B 40 C

there are two, GF and DE, each with a weighting of 38 30
18 52
20. It does not matter which of these two we choose. 22
We will choose DE, as shown. 25 G D
A 36 H
20 35 24 20
34
E
F 75
Continuing this process sees us choose GF (20) then B 40 C

HC (22) then HD (24) and then AG (25) as our next 30
38 18 52
four edges, as shown on the right.
22
25 G D
A 36 H
20 35 24 20
34
E
F 75
The next ‘shortest’ edge will be CD (30). However this B 40 C

will not bring any vertex on line that is not already on 30
38 18 52
line. (It will give an unnecessary cycle.)
22
25 G D
The next ‘shortest’ after CD is AF (34) which again A 36 H
20 35 24 20
introduces an unnecessary cycle. 34
E
Choosing the next ‘shortest’ which does not introduce F 75
a cycle we choose FH (35), and so the minimum

spanning tree is completed.

Exercise 5A
Find the minimum spanning tree for each of numbers 1 to 6, and state its length in each case.
1 B 94
G
26 40
30 E 20
33
40 C 50
A H
26
40
30
23
52
50
F
D
2 C 400 G
220
B 500
150
200 250 F 600
300 I
A 270
400 D 200 370
280 H
250
E
3 30
20
14 22
30
11
23
30
12
17 15
20 15
14 25
19
4 20 16
5
30 14 19
23
15
8 18
9
10
14
12 18 17
13

5 F 20
H
18
45
G 40
40
40
A 42 C 22
30 E 45 I
25 35 16
36 60
B 50 D
6 320
220 700
320
250
270
300 420
300
360 150
550
200
330
240
200
7 A market gardener wishes to replace the old piping currently connecting his seven-outlet
irrigation system. Rather than replacing all of the existing piping he wishes to replace the system
with the minimum total length of piping that will keep all seven outlets connected, either directly
or indirectly. He will then close off those unnecessary parts of the existing system.
The graph on the right shows the existing G 100 F 55 E
system. Points A to G are the outlets and the
numbers indicate the length of each section 40 45
50 120 65
in metres.
C 80
115 D
Which sections should he replace with new A
piping, which should he close off, and what 35 90
120
total length of new piping will be needed?
B

8 A camp-site is being designed with three ablution blocks, A1, A2 and A3, two play areas for children,
P1 and P2, two barbecue areas, B1 and B 2, and a shop, S. The sketch below shows the relative
locations of these features with the distances between them given in metres.
65 A2 55 B1 50
P2 P1
70
75 35 30
40 40
60 50
A1 A3
B2
40 60
70
Paths are to be built so that all of these features are linked. On a copy of the above diagram
indicate the network of paths that spans all of the features whilst minimising the total length
of pathway.
What is this minimum length of pathway?
9 The network below shows the system of tunnels that exist linking the seven entrances of an old
silver mine. The numbers on the edges show the length of each section of tunnel in metres.
C 80 D
30
50 75 55
B 45
100
E
eseorfynnad/moc.kcotSi
G
40 60
55
65
80
A
F
The mine is a popular tourist attraction but all of the tunnels are in need of repair if they are
to remain open for visitors.
The owners cannot afford to repair all of the tunnels so instead decide to repair the minimum
total length possible, while still allowing entrance to the repaired system to be possible from all
seven of the existing entrances.
The tunnels not being repaired will then be closed off to visitors.
They ask you to determine which tunnels should be repaired and which shut down.
a On a copy of the network show the tunnels the owners should repair.
b What is the total length of tunneling that will be closed off to visitors?

10 The table below shows the distances, in km, along the roads that exist between seven towns:
Alsop, Beal, Clan, Dawse, Erle, Felix, and Grant.
(A ‘–’ indicates that there is no direct road between those towns.)
Alsop Beal Clan Dawse Erle Felix Grant

Alsop – – 310 – 300 – 620
Beal – – 550 430 530 330 –
Clan 310 550 – – 350 – –
Dawse – 430 – – – 210 390
Erle 300 530 350 – – 380 360
Felix – 330 – 210 380 – –
Grant 620 – – 390 360 – –
The sketch on the right shows the approximate

Clan Beal
positions of the seven towns with respect to
each other.
Alsop
a Make a copy of this sketch and show on
Erle
your copy the roads that exist between
the towns and the lengths of these roads.
Felix
b Water pipes are to be laid along some of Grant Dawse
these roads so that the seven towns are, either
directly or indirectly, connected to each other.
Determine the minimum length of piping required.
Minimum spanning tree from the distances table

In the last question of the previous exercise it was reasonably easy to transfer the details given in the
table onto a network diagram because we were given a diagram showing the approximate locations of
the towns with respect to each other. We then used the network diagram to determine the minimum
spanning tree. However, if the network involved had been larger, and/or the relative locations of the
vertices not have been known, the task of creating the network diagram with all of the weightings in
place would have been more difficult. Fortunately it is possible to determine the minimum spanning
tree directly from the table of distances without having to create the network diagram first. This
process is demonstrated on the next page. However, as you read through it you should notice that the
method is actually Prim’s algorithm …
choose any vertex as our initial ‘on line’ vertex and then build up the spanning tree by connecting on line
vertices to the ‘nearest neighbour’ whilst always making sure that no cycles are introduced
applied to the table of distances rather than to the network drawing.

Consider the table of distances shown on the right, A B C D E F G
which is actually the one from the last question of
A – – 310 – 300 – 620
the previous exercise.
B – – 550 430 530 330 –
C 310 550 – – 350 – –
D – 430 – – – 210 390
E 300 530 350 – – 380 360
F – 330 – 210 380 – –
G 620 – – 390 360 – –
Whilst we can start at any vertex we will choose

↓
to start at A.
A B C D E F G
With A on line we look down the A column to see – – 310 – 300 – 620
A
which is the shortest connection that can be made
B – – 550 430 530 330 –
from A.
C 310 550 – – 350 – –
In this case it is the 300 km connection to town E.
D – 430 – – – 210 390
E 300 530 350 – – 380 360
F – 330 – 210 380 – –
G 620 – – 390 360 – –
Having chosen to look down the A column for a

↓
connection from A we then rule a line through the
A B C D E F G
A row to ensure that we do not, at some later stage,
use this line to make an unnecessary connection A – – 310 – 300 – 620
back to A. B – – 550 430 530 330 –
(Alternatively, we could have chosen to look across C 310 550 – – 350 – –

row A for the shortest connection, and then rule a D – 430 – – – 210 390
line through column A.) E 300 530 350 – – 380 360
F – 330 – 210 380 – –
G 620 – – 390 360 – –
Having selected the 300 km connection to vertex E

↓ ↓
we now have both A and E on line and indicate this
by the arrows at the top of these two columns. A B C D E F G
A – – 310 – 300 – 620
Again, to avoid making unnecessary connections at
B – – 550 430 530 330 –
a later stage we rule a line through the E row .
C 310 550 – – 350 – –
We now look down our two on line columns, A and
D – 430 – – – 210 390
E, and choose the minimum connection that can be
made from one of these two vertices. E 300 530 350 – – 380 360
F – 330 – 210 380 – –
G 620 – – 390 360 – –

↓ ↓ ↓
This gives us the 310 km road from A to C as our A B C D E F G

next connection.
A – – 310 – 300 – 620
This brings vertex C on line and we rule a line B – – 550 430 530 330 –
through the C row. 310 550 – – 350 – –
C
D – 430 – – – 210 390
E 300 530 350 – – 380 360
F – 330 – 210 380 – –
G 620 – – 390 360 – –
We now look down our three on line columns,
↓ ↓ ↓
A, C and E, and choose the minimum connection
A B C D E F G
that can be made from one of these three vertices.
A – – 310 – 300 – 620
This gives us the 360 km road from E to G as
B – – 550 430 530 330 –
our next connection.
C 310 550 – – 350 – –
D – 430 – – – 210 390
E 300 530 350 – – 380 360
F – 330 – 210 380 – –
G 620 – – 390 360 – –
This brings town G on line and we rule a line

↓ ↓ ↓ ↓
through the G row.
A B C D E F G
We continue this process until all of the towns – – 310 – 300 – 620
A
are on line. The circled items in the table
B – – 550 430 530 330 –
indicate the direct connections that together
form the minimum spanning tree. C 310 550 – – 350 – –
D – 430 – – – 210 390
E 300 530 350 – – 380 360
F – 330 – 210 380 – –
G 620 – – 390 360 – –
From the completed table on the right we see that
↓ ↓ ↓ ↓ ↓ ↓ ↓
the roads AC, AE, EF, EG, FB and FD form the
A B C D E F G
minimum spanning tree, with a total length
of 1890 km. A – – 310 – 300 – 620
B – – 550 430 530 330 –
The reader should confirm that this is the same
answer as was obtained for question 10 of C 310 550 – – 350 – –
Exercise 5A. D – 430 – – – 210 390
This application of Prim’s algorithm may at first E 300 530 350 – – 380 360
seem rather tedious and lengthy but you will find F – 330 – 210 380 – –
that it can be carried out quite quickly and, G 620 – – 390 360 – –
of course, in practice it only requires one copy
of the table. The multiple copies shown here were
to allow each step of the process to be clearly seen.

Exercise 5B
Determine the length of the minimum spanning trees for the networks defined by the tables in
questions 1 to 5. (A ‘–’ in the table indicates there is no direct route between the locations.)
1 A B C D
2 A B C D E
A – – 21 20 A – 38 – 62 32
B – – 39 35 B 38 – 37 – 35
C 21 39 – 24 C – 37 – 40 22
D 20 35 24 – D 62 – 40 – 41
E 32 35 22 41 –
3 A B C D E F
4 A B C D E F G
A – 450 – – – 470 A – 2.6 – – – 3.6 4.0
B 450 – 300 – – 400 B 2.6 – 5.0 – – – 2.8
C – 300 – 500 200 400 C – 5.0 – 5.0 3.0 – 3.5
D – – 500 – 420 700 D – – 5.0 – 5.3 – –
E – – 200 420 – 310 E – – 3.0 5.3 – 5.7 3.9
F 470 400 400 700 310 – F 3.6 – – – 5.7 – 2.8
G 4.0 2.8 3.5 – 3.9 2.8 –
5 A B C D E F G H I
A – 650 – – – 370 400 – –
B 650 – 600 – – – 340 350 –
C – 600 – 220 – – – – 380
D – – 220 – 560 – – – 390
E – – – 560 – 700 – 300 210
F 370 – – – 700 – – 410 –
G 400 340 – – – – – – –
H – 350 – – 300 410 – – 250
I – – 380 390 210 – – 250 –
6 Determine the length of the minimum spanning A B C D E F

tree for the network defined by the table on the
A – 8.7 – 8.9 – 9.1
right. (A ‘–’ in the table indicates there is no direct
route between the locations.) B 8.7 – 9.5 – – 4.5
Name the five roads that would be used in your C – 9.5 – 6.1 7.0 5.3
minimum spanning tree. D 8.9 – 6.1 – 3.5 –
E – – 7.0 3.5 – 3.8
F 9.1 4.5 5.3 – 3.8 –

7 Eight farmlets (A, B, C, D, E, F, G and H) are to be connected to mains water supply. The water
authority plans to lay pipes alongside some of the existing roads that link the farmlets. The plan
is to bring all eight farmlets ‘on line’ using those roads that form the minimum spanning tree for
the network of existing roads. In this case ‘minimum’ meaning the minimum total length. The
lengths of the existing roads are as in the table below. All lengths are given in kilometres and
a ‘–’ in a space means that there is no existing road directly linking those two places.
Determine the minimum length of pipe required.
A B C D E F G H
A – 4.2 – – 7.2 4.1 – –
B 4.2 – 6.0 – – 2.0 – –
C – 6.0 – 4.4 – 6.3 3.0 –
D – – 4.4 – 6.8 – 5.0 6.0
E 7.2 – – 6.8 – – – 2.3
XFLOG/moc.kcotSi
F 4.1 2.0 6.3 – – – 3.5 2.8
G – – 3.0 5.0 – 3.5 – 2.5
H – – – 6.0 2.3 2.8 2.5 –
8 A school has six teaching areas, A, B, C, D, E and F, A B C D E F

that need to be linked for the school video, public
A 0 $170 $525 $410 $210 $325
address, internet and phone systems.
B $170 0 $400 $310 $145 $225
Cables are to be installed to create a spanning tree
for the network with vertices at A, B, C, D, E and F. C $525 $400 0 $140 $315 $200
Find the minimum cost of the spanning tree if the D $410 $310 $140 0 $190 $100
cost of the direct cable links between vertices are E $210 $145 $315 $190 0 $115
as in the table on the right. F $325 $225 $200 $100 $115 0
kesuzi/moc.kcotSi

9 A railway authority wishes to build a freight railway network linking seven towns. The network is
to be constructed so that it will be possible to travel by rail from any one of the towns to any of the
other six in the network, perhaps not directly, but at least by going via other towns in the network.
The towns are Harlow, Gatley, Olber, Wayley, Raine, Lewis and Fitch and the distances involved
for the feasible rail routes from any of these towns to any of the others are given in the following
table (in kilometres).
Harlow Gatley Olber Wayley Raine Lewis Fitch

Harlow 0 300 510 316 122 195 320
Gatley 300 0 224 100 212 378 206
Olber 510 224 0 200 400 517 269
Wayley 316 100 200 0 202 327 112
Raine 122 212 400 202 0 189 198
Lewis 195 378 517 327 189 0 252
Fitch 320 206 269 112 198 252 0
a Based on these distances determine the minimum length of track required for this network.
b In this ‘minimum length network’ which six pairs of towns would have a direct rail link?
(By direct rail link we mean able to journey from one town to the other without having
to pass through any of the other towns on the way.)
10 The table below indicates the distances, in kilometres, along roads that exist between ten locations,
A to J. (A dark shaded cell in the table indicates that there is no direct road between the locations.)
A B C D E F G H I J
A 60 30 70
B 45 20 40
C 65 28 35 30
D 60 65 46 18 20 43
E 30 45 46 42
F 20 18 42 41
G 70 20 38
H 28 43 38 54
I 35 54
J 40 30 41
The roads are in need of upgrading and it is decided that, rather than upgrade all of the roads, only
those roads needed to ensure that all ten locations are linked by upgraded roads, either directly or
indirectly, will be upgraded.
Which roads should be upgraded if the total length of road to be upgraded is to be kept to
a minimum and what would this minimum length be?

Miscellaneous exercise five
1 The following employment data gives the number of people in full-time employment in Australia
each quarter from January 2010 to April 2013, with the exception of July 2011 for which the
information is missing from the table.
Data point (n) 1 2 3 4 5 6 7

Month Jan April July Oct Jan April July
2010 2010 2010 2010 2011 2011 2011
Employed (’000) (N) 7630 7677 7749 7821 7876 7897
Data point (n) 8 9 10 11 12 13 14

Month Oct Jan April July Oct Jan April
2011 2012 2012 2012 2012 2013 2013
Employed (’000) (N) 7911 7942 7971 7989 8024 8028 8022
(Based on Australian Bureau of Statistics data.)
a Use linear regression techniques to determine the equation of the least squares line of best fit,
N = an + b.
b Interpret your value of a in your answer for part a.
c Use your regression line to predict the employment figure for July 2011 and for January 2015
stating which answer you would expect to be the more reliable and why.
d View the given data on a graphic calculator or computer and write some appropriate comments.
2 In an attempt to determine how long an investment of $350 000 placed in an account paying
6% per annum, compounded monthly, would last if $1500 was withdrawn after one month and
at the end of every month thereafter, Jim put this information into a financial calculator, as
shown below left. When he asked the calculator to determine N, the number of withdrawals the
calculator returned ‘Error’, as shown below right. Explain.
Compound Interest Compound Interest

N N Error
I% 6 I% 6
PV –350000 PV –350000
PMT 1500 PMT 1500
FV 0 FV 0
P/Y 12 P/Y 12
C/Y 12 C/Y 12

3 The table below shows some of the data for the number of kilolitres of water used by a household
each quarter for a period of three years.
t Year Quarter No. of kL Uncentred Centred

4-pt MA 4-pt MA
1 One 1st 95 –
–
2 2nd 60 –
a
3 3rd 36 d
b
4 4th 77 78
c
5 Two 1st 131 e
82
6 2nd f 81
g
7 3rd 44 79
h
8 4th i k
j
9 Three 1st 123 m
l
10 2nd 92 84
n
11 3rd 60 –
–
12 4th o –
Calculate the value of a, b, c, d, … o.
4 To the nearest cent, what ‘one-off’ lump-sum payment must be invested into an account paying
compound interest of 8% per annum, compounded monthly, for the investment to be worth
$5000 four years later?
5 Amy takes out a loan of $15 000. Interest of 8.2% of the outstanding balance is added at the end
of each year and then Amy’s regular annual repayment of $2500 is taken from the outstanding
balance.
a How much does Amy still owe on this loan immediately after she makes the $2500 repayment
at the end of year six?
b If instead Amy wishes to pay the loan off in the six years what does her regular annual
repayment need to be?

6.
Maximum flow
• Maximum flow – a systematic approach
• Maximum flow = minimum cut
• Miscellaneous exercise six
Situation
Suppose the network on the right shows the arrangement C
of water pipes around a farm with the numbers indicating
the maximum rate of flow that each pipe can manage, in 100 120
litres per minute.
100 B 50
A
Point A is the source of the water supply. The three pipes Z
coming from A could draw water from A at a total rate of 150 200
D
350 litres every minute but could this amount of water be
delivered to Z each minute?
What is the maximum amount of water that can be delivered to Z each minute?
Now suppose we add a new pipe with a capacity of 50 litres per minute. The networks below show two
ways that this could be done. What is the maximum flow from A, the source, to Z, the sink, in each of
these situations?
C C
100 120 100 120

50
B
100 B 50 100 B 50
A A
Z Z
50
150 200 150 200
D D
When the network is reasonably simple we can determine the maximum flow by mental reasoning
as we have done above. However, for more complicated networks we need a procedure, or algorithm,
to follow. Before such an algorithm is suggested for you, try to develop a procedure for yourself to
determine the maximum flow through the network below from a source at A to a sink at G. The
network shows the units of electrical power that can be delivered along the various links between
relay stations.
200
C F
250 100 150 350

50
150 90 50
A B E G
50
200 100
D
Note: In reality, in situations like these there is often more than one source from which we can obtain
the water or the power or whatever it is that is involved. In addition there is frequently more
than one sink where this water or power is required. However, in this book we will restrict our
attention to situations in which there is just one source and just one sink.
ISBN 9780170395069 6. Maximum flow 109

Maximum flow – a systematic approach
Suppose we wish to determine the maximum flow that can be achieved for the network shown below
from the source at A to the sink at H.
C 280 F
450
400 200 G 600 H
200
300 B 150 E
A
200 150 150
100
To approach the problem systematically we consider the separate routes in turn, from the top of the
diagram to the bottom. We send the maximum flow we can along each route and note the remaining
capacity of each pipe involved.
Consider route ACFGH. C 280 F
450
This route can carry a maximum of 280 400 200 G 600 H
units (because that is all that CF can carry). B E
200
A 300 150
200 150 150

100
Sending 280 units along ACFGH C 280 0 F 170

leaves each pipe involved with the carrying 120 450 320
capacity indicated in the next diagram. 400 200 G 600 H
200
(Check carefully that you agree with these 300 B 150 E
A
figures and can see how they are obtained.)
200 150 150
100
Now consider ABEFGH. 20

C 280 0 F 170 170
We can send 150 units along this route. 120 50 450 320
400 200 G 600 H
The remaining capacity of each pipe 150 0 200
300 B 150 E
will then be as shown on the right. A
200 150 150
100
We can send 20 units along ABDEFGH. 0

20
C 280 0 F 170
150
30 170
120 450 320
400 130 50
200 G 600 H
150 0 200
300 B 150 E
A
180 130
200 150 150
100

0 20
20
We can send 130 units along ABDEGH. C 280 0 F 170
150
30 170
120 450 320
400 130 0 50
70 G 600 H
200
150 0 200
300 B 150 E
A 0
180 50
200 130 150
100 150
D
0 20
20
We can send 100 units along ADH. C 280 0 F 170
150
30 170
120 450 320
400 130 0 50
70 G 600 H
200
150 0 200
300 B 150 E
A
180 50 0
200 130 50
100 0 150 150
Thus the flow is: ACFGH 280 units +
ABEFGH 150 units

ABDEFGH 20 units
ABDEGH 130 units
ADH 100 units
giving a total of 680 units
The maximum flow from the source at A to the sink at H is 680 units.
Note: • This process would normally be carried out using a single diagram as shown in the example
on the next page.
• The arcs leaving A have a total capacity of 800 units ( = 400 + 300 + 100). Our maximum
flow of 680 would indicate that there must be 120 units that we have not been able to use.
This can be seen in the diagram above as 120 units of unused capacity in AC.
Similarly the arcs going to the sink, H, have the capacity to deliver 750 units ( = 600 + 150).
Our maximum flow would indicate there must be 70 units that we have not been able
to use. This can be seen in the diagram above as 50 units of unused capacity in DH and
20 units of unused capacity in GH.
• By noting the unused capacity in each arc we can determine the flow along each pipe
to obtain maximum flow:
C 280 F
450
280 170 G 580 H
130
300 B 150 E
A
150 150 100
100
Notice in this diagram that 680 units (the maximum flow) leaves A, 680 units arrives at H
and at all of the other vertices the total number of units entering is the same as the total
number leaving.

• This diagram showing how the maximum flow can be achieved is not necessarily unique.
If we use this method of considering the separate routes but start from the bottom route,
ADH, we can arrive at the following arrangement for delivering the maximum flow.
Note that the maximum flow is still 680 units but the way it is achieved is different.
C 280 F
330
280 50 G 530 H
200
300 B 100 E
A
200 150 150
100
D
• Considering separate routes, one after another, will not necessarily guarantee that we obtain
the maximum flow.
For example, in the network on the right, if we were to B
consider route ABDC first and direct 40 units along 40 50
40
this route, we obtain a total flow of just 40 units,
A C
which is clearly not the maximum. To direct 40 units 40 D 40
along ABDC in this way is clearly not the wisest thing
to do but in some more complicated networks the ‘wisest’ strategies may not be so obvious.
However, the method outlined here, with a little common sense applied, should give the
maximum flow for most networks you are likely to meet in this course. Later in this chapter,
we will see a way of checking our value to see if it is the maximum flow.
EXAMPLE 1
Determine the maximum flow possible for the B 20 E

network of pipes shown on the right with point A 30 30 40
as the source and point F the sink. 20
25 C 10
Redraw the network showing the flow you would A F
10
direct along each pipe in order to achieve the 25 D 40
maximum flow at F.
Solution
Using the systematic approach: 0 B 20 0 E 0
10 10
ABEF: 20 units 20
30 30 20 10 0 40
ABCEF: 10 units 20
25 15 5 0 C 10 0
ACEF: 10 units A F
10 5 10
ACF: 10 units 25 0 40
35
D
ACDF: 5 units
ADF: 25 units Total: 80 units
The maximum flow from A to F is 80 units B 20 E
and can be achieved as follows: 30 10 40
20
25 C 10
A F
5
25 D 30

Exercise 6A
For each of the networks shown in questions 1 to 4 write down the maximum flow from the source at
A to the sink at X (any working should be done mentally).
1 100 80 60 2 B
A B C X 40 30
A X
60
3 B 4 B
50 40
40 50
30 10 40 X
A
A X C
60 40 30 20
C
D
For each of the networks shown in questions 5 to 10 use a systematic approach to determine the
maximum flow from the source at A to the sink at Z.
5 C 350 D 6 B 150 E
300 400 200 200

100
100 40 170
A 100 Z
A C Z
200 100
150 400
250
50 E
B D
7 C 8 D
200 250 60 100

10
100 D 150 C
A Z A Z
20 30
100
400 150 100 40 30 90
B
B 50 E
9 B 10 E 30 B
250 100 10 55
10 30
100
250 100
Z
A C Z 100 F A
E 300
100 50
250 250 65
D 200 F D 70 C

11 For the network shown on the right determine Q U
200
the maximum flow possible from the source at
P to the sink at V. 400 150 200
T
Redraw the network showing the flow you would 250
150 S 50
P V
direct along each arc in order to achieve the
250
maximum flow at V. 350 200
400
12 Given that the network shown on A D

150 x F
the right shows the flow along each
50
arc that would achieve the 200 y
maximum flow from the source at w 300
450 C E
A to the sink at G, determine the v z
values of v, w, x, y and z.
B 350 G
13 The network shown on the right represents B 1100 D

a telecommunications system linking five
locations, A to E. The number on each 200
800 200 1300
edge represents the maximum number of
connections that can be active along that 700 C 200
edge simultaneously. What is the maximum
number of simultaneous connections that A 400 E
can exist from location A to location E?
Draw a diagram showing the number of connections you would make along each edge, and show
the directions of those connections, in order to produce this maximum situation.
14 The network shown below represents a system of water pipes connected to a main supply at A.
The arcs in the network represent the pipes in the system and the numbers on the arcs give the
maximum capacity of that section of pipe in units per minute.
B 6 E
7 8
5 2
A G
4 7
D
5 6 4 5
C 6 F
What is the maximum amount of water that could be delivered at point G per minute given that
the only restriction is the capacity of the pipes?
One of the three 5-units-per-minute pipes is to be upgraded to an 8-units-per-minute pipe.
What would be the effect of each of these three possible upgrades on the maximum flow at G?

Maximum flow = Minimum cut
Earlier it was shown that by choosing an inappropriate route to start with, the total flow we end up
with may not be the maximum flow possible. If the network is very complicated, mistakes can easily be
made. With careful use of our systematic method it will give us the maximum flow for the networks
we are likely to meet in this course but we can check that it is the maximum flow by considering ‘cuts’.
Consider the network shown C 280 F
on the right. 450
400 200 G 600
200 H
300 B 150 E
A
200 150
100 150
First draw vertical lines through the source on the left and through the sink on the right. A cut is then
made by starting above the network and drawing a line that cuts only arcs of the network and finishes
below the network. (The cut disconnects source from sink.)
The value, or capacity, of the cut I II
C 280 F
is found by summing the cut arcs.
450
400 200
The diagram on the right shows two G 600
200 H
such cuts, I and II, and their values. 300 B 150 E
A
The next diagram shows two more cuts, 200 150
100 150
III and IV.
D
Note that IV cuts EF from ‘below to 800 730
above’ (imagine EF horizontal with
its direction of flow being from left to III IV
C 280 F
right, i.e. source to sink). This is not 450
included when determining the value 400 200 G 600
200 H
of the cut. (The value comes from those 300 B 150 E
arcs that are cut from ‘above to below’.) A
200 150
100 150
We can use the fact (not proved here)
that the maximum flow equals the 900
730 D
minimum cut to check our maximum
flow. A cut value and a flow value can
V
only be the same when the cut value C 280 F
is a ‘min’ and the flow value a ‘max’. 450
400 200
Hence, if we can find a cut equal to G 600
200 H
our maximum flow then our value 300 B 150 E
A
must indeed be the maximum. 200
100 150 150
When we met this network earlier
we calculated the maximum flow as 680 D
680 units. We can find a cut having

this value (see diagram) thus confirming
our maximum flow value.

Exercise 6B
1 Determine the value of the I II V III IV
B 30 F
cuts I to V shown in the
diagram on the right. 50 50
30 40
A G
30 D 40
100 50
C 50 E
The networks below were all encountered in Exercise 6A. For each network confirm the stated
maximum flow by finding a cut of the same value.
2 C 350 D 3 B 150 E
300 200
400 200
100
A 100 100 40 Z 170
A C Z
200 100
150
250 400
50 E
B D
Maximum flow 450 units Maximum flow 550 units
4 C 5 D
200 250 60 100

10
100 D 150 C
A Z A Z
20 30
100
40 30
400 150 100 90
B
B 50 E
Maximum flow 500 units
6 B 7 E 30 B
250 100 10 55
100 10 30
250 C 100
Z
100 F A
A E Z
300
100 50 65
250 250
D 70 C
D 200 F

Miscellaneous exercise six
1 Apply the average (mean) percentage method explained in Chapter 2 (p. 33) to the data given
below to determine the seasonal index for each quarter, Q1, Q2, Q3 and Q4, and explain what it
is that these indices tell us. Give the indices as percentages rounded to the nearest whole percent.
Year 1 Year 2
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
688 498 248 406 733 632 352 483
2 Find the length of the minimum A B C D E F G H

spanning tree for the network defined
A – 37 70 68 – 80 – –
by the table on the right.
B 37 – 40 – – – 58 –
(A ‘–’ in the table indicates there is no
direct route between the locations.) C 70 40 – – – – 52 –
D 68 – – – 17 – 36 –
E – – – 17 – 20 – 67
F 80 – – – 20 – – 78
G – 58 52 36 – – – 39
H – – – – 67 78 39 –
3 (Use a calculator with a built-in facility to calculate compound interest calculations or a suitable
online finance calculator.)
A loan of $500 000 is taken out to finance the purchase of a house. Interest is charged at 8.5% per
annum with interest compounded monthly.
a How much must be repaid per month if the loan is to be paid off in 25 years?
b What is the total amount of interest paid in the 25 years? (Answer to the nearest $10.)
4 A ‘one-off’ investment to open a savings account, in which interest is calculated annually, is such that
if n years later the value of the account is $Tn then the value one year after that, $Tn 1, is such that +
Tn + 1 = Tn × 1.072 and T 0 = 12 000.

a What is the initial investment?
b What is the annual interest rate?
c Using a recursive approach and not the financial capability of some calculators and computer
programs, determine the value of the account after six years.
5 Under a reducing balance (fixed percentage) depreciation scheme of 12% per annum, how much
does an asset initially worth $120 000 depreciate in each of its first three years?

6 The diagram on the right shows the pipe system B 200 C
connecting a number of watering points on
a farm to the main source, A. The number 300
400 200
on each section indicates the maximum flow
700 900
that pipe can manage in litres per minute.
What is the maximum number of litres per 200 D E 100
250
minute that could be delivered at outlet
A F
F under this system? Make a copy of the 500
diagram and indicate the flow along each
pipe (in L/min) to achieve this maximum delivery at F.
7 The following table shows the number of short-term visitor arrivals into Australia for each four-
month period from Jan-Feb-Mar-Apr 2008 to Sept-Oct-Nov-Dec 2013.
n Months Year Number of short-term visitors 3-pt MA (M)
1 Jan-Feb-Mar-Apr 2008 1 898 800 –
2 May-Jun-Jul-Aug 2008 1 691 100
3 Sept-Oct-Nov-Dec 2008 1 922 500
4 Jan-Feb-Mar-Apr 2009 1 872 200
5 May-Jun-Jul-Aug 2009 1 587 200
6 Sept-Oct-Nov-Dec 2009 2 030 800
7 Jan-Feb-Mar-Apr 2010 1 933 600
8 May-Jun-Jul-Aug 2010 1 720 400
9 Sept-Oct-Nov-Dec 2010 2 136 100
10 Jan-Feb-Mar-Apr 2011 1 959 800
11 May-Jun-Jul-Aug 2011 1 717 700
12 Sept-Oct-Nov-Dec 2011 2 093 300
13 Jan-Feb-Mar-Apr 2012 2 023 100
14 May-Jun-Jul-Aug 2012 1 777 600
15 Sept-Oct-Nov-Dec 2012 2 231 500
16 Jan-Feb-Mar-Apr 2013 2 103 100
17 May-Jun-Jul-Aug 2013 1 895 700
18 Sept-Oct-Nov-Dec 2013 2 381 700 –
With the assistance of a computer spreadsheet, or otherwise

a Determine the entries for the final column, the 3-point moving averages, giving values
to the nearest hundred.
b Produce a graph with n on the horizontal axis and displaying both the raw data and the
moving average data as time series line graphs.
c Use linear regression to obtain the equation of the line of best fit for predicting M , the
moving average, given n, i.e. the equation M = an + b . (Give a to the nearest 10 and b to
the nearest 1000.)
d Interpret the value of ‘a’ in the equation from part c.

7.
Project networks
• Project networks
• Constructing a project network
• Miscellaneous exercise seven
Situation
A company is contracted to carry out a particular job. The company splits the job into ten smaller tasks
and sub-contracts each of these to other companies and individuals. The main company then monitors
the progress of the job by attempting to keep it to the program shown in the network below.
B4
I6
A5
C4 H4
D6 G1 J3
E7
F2
• In this network the ten tasks are labelled A to J and are represented by the directed edges, or arcs.
The number on each arc shows the time in days that is allowed for the task.
• The network indicates any task that must be completed before another can commence. Thus
task H, for example, has 4 days allowed for it and requires task B to be completed before it can be
commenced (and task B required A to be completed before it could be commenced). Provided these
‘prior task commitments’ are met other tasks can be going on at the same time. For example tasks A,
D and E can all be going on at once.
What is the least number of days required for the job to be completed according to the time allocations
on the above schedule?
Poor weather may cause task C to take 6 days rather than 4. If this occurs what will be the least number
of days for the job now?
If task I were to be completed in 5 days rather than 6, and all others were to be completed using the
allotted time, what will be the least number of days for the job now?
If task J were to be completed in 1 day rather than 3, and all others were to be completed using the
allotted time, what will be the least number of days for the job now?
llewnraB luaP/moc.kcotsrettuhS
ISBN 9780170395069 7. Project networks 121

WS
Project networks
The situation on the previous page involved a project network. How did you get on? Did you develop
Calculating EST
and LST
a systematic approach? Did you manage to think it through and answer the questions or did all the
things you had to consider leave you confused?
One systematic approach involves two stages:
WS
1. Forward scan. I.e. follow the network in the direction of the arrows and at each vertex record the
Critical paths, critical
times and activity
least time needed to reach this stage in the project with all prior tasks completed. In this way we
float times consider the earliest start time (EST) of an activity.
2. Backward scan. I.e. backtrack through the network recording the latest time each vertex could be
reached without delaying the minimum completion time. In this way we consider the latest start
time (LST) of an activity.
This approach is demonstrated in the following example.
EXAMPLE 1
Find the minimum completion time and the critical path for the project network shown below.
The arcs P, Q, …, Y represent the tasks involved and the numbers on the arcs indicate the time in
hours that the task requires.
Q8 U3 W9 V18
R8
P6 X4 Y6
S12
T9
What is the maximum number of extra hours that could be allocated to task S, over and above the
12 in the network, without extending the minimum completion time?
Solution
Work through the network from the start and at each vertex record the least time needed to reach
that stage in the project with all prior tasks completed. Check that you understand the placement
of the numbers in the circles below.
Earliest this stage can be Earliest this stage can be reached is
reached is 6 hours into 14 hours into the project, to allow
the project, to allow task prior tasks (Q and P) to be finished.
P to be finished. 14
Q8 U3 W9 V18
R8
0 6 18 23 32
P6 X4 Y6
S12
T9
Earliest this stage can be reached is 18 hours into the project, to allow prior tasks U, R and S
(and hence Q and P as well) to be finished. The critical path to this stage will be P-S because
that requires the full 18 hours. Other paths require less and thus have some ‘room for delay’
(P-R requires 14, P-Q-U requires 17). This ‘room for delay’ is known as float or slack.
Thus the minimum completion time is 32 hours.

Now, to identify the slackness in the system, work back through the network and at each vertex
record in the lower half of the circle the latest time we could reach that stage whilst still being
able to complete the project in the minimum completion time. Check that you understand the
placement of these numbers in the bottom halves of the circles below.
14 (= 32 – 18) is the latest time that we can reach this stage and still
complete the project in 32 hours, the minimum completion time.
Earliest this stage can be reached with all prior tasks complete is 14 hours
into the project. Thus there is no slackness available here.
14
Q8 14
U3 W9 V18
R8
0 6 18 23 32
0 P6 6 22 X4 26 Y6 32
S12
T9
26 (= 32 – 6) is the latest time that we can reach this stage and still
complete the project in 32 hours, the minimum completion time.
Earliest this stage can be reached with all prior tasks complete is 23 hours
into the project. Thus there is some slackness available here.
Any vertex for which the earliest time the vertex can be reached is the same as the latest time the
vertex can be reached lies on the critical path. If we journey from start to finish, through such
vertices we can obtain the critical path, P-Q-V. If any of these tasks are delayed the minimum
completion time will be extended.
Consider task S and notice that this task will be ready to start 6 hours into the project and does not
have to finish until 22 hours into the project. Thus we could allow 16 hours for this 12-hour task.
Thus we could allocate an extra 4 hours for task S, over and above the 12 in the network. Task S is
said to have a float, or slack time of 4 hours.
Consider task X. We say that X has an earliest start time (EST) of 18 hours and a latest start time
(LST) of 22 hours (26 - 4). X has 4 hours float.
Consider task W. It has an earliest start time (EST) of 14 hours and a latest start time (LST)
of 17 hours (26 - 9). W has 3 hours float.
All activities on the critical path have zero float time.
Note: • In the above example the diagram is drawn twice and comments are included in boxes.
The reader would not have to include such comments and would only need one diagram.
• The float, or slack, considered here is the time by which an activity can be delayed, from
its early start time, without altering the project’s minimum completion time. It is sometimes
referred to as the total float of an activity.

B8
• Consider the diagram:
0 A20 20 30 D8 38
0 20 30 38
C10
The numbers in the circles may at first glance give the impression that there is no float, or
slack, in the system but with two arcs from the second vertex to the third vertex the critical
path is determined by taking task C, the task that takes the greater time. Thus A - C - D is
the critical path. Task B has 2 units of float, a fact that can be missed if we only look at the
numbers in the circles.
Exercise 7A
1 The times for tasks P to Z shown in the project network below are in days.
R4
S4 X4
Q5
P7 W3 Y1
T6 U7
Z4
V4
a Determine the minimum completion time.

b Determine the critical path.
c Task V is a task that is expected to take 4 days. Assuming all the other tasks can be completed
in exactly the stated times, how many extra days could be allocated to task V without altering
the minimum completion time?
2 The times for tasks A to H shown in the project network below are in hours.
B7
A4 C3 G3
D1
E1
F2 H5

c Some problems are experienced with task D and it requires an extra 3 hours. What is the
minimum completion time for the project now?
d How many extra hours could be allowed for task B, over and above the 7 hours allocated
on the above network, without extending the minimum completion time?

3 The times for the tasks shown in the project network below are in days.
B7
C5 J3
A3 K3
E4 H4
N2
D2 F3 L3
G4 M5

c Due to poor weather, tasks D and B each suffer a two day delay in commencement.
What is the minimum completion time now?
4 The times for the tasks P to Z shown in the project network below are in hours.
Q7 R5
T6
P6
S5 U 13
Z7
W 11 V9
Y2
X7

c The project managers can employ extra staff and reduce the time required for task V by
3 hours. If they take this option will the minimum completion time be cut by 3 hours?
d Instead of the option outlined in part c the managers decide to allocate extra staff to one
particular task and this reduces the time required for that task by 2 hours and it cuts the
minimum completion time by 2 hours. Which task did they allocate the extra staff to?

5 The times shown in the project network below are in days.
B7 H8
C9
A7 G8 J6
D6
E 10 F 10

c What is the earliest start time (EST) for task H?
d What is the latest start time (LST) for task H if the minimum completion time is to be
unchanged from that given in the answer to part a?
e What is the float time for task H?
f What is the float time for task F?
g What is the float time for task D?
h Due to staff sickness it is necessary to reallocate the staff involved on tasks D and F. This will
cause one of these jobs to require 5 extra days to complete. Which of the two tasks should be the
one chosen to require this extra time if the completion time is to be kept as low as possible?
6 The times for tasks P to Z shown in the project network below are in hours.
S 25
R 12
P 20 Q 17
U 18 W 17
T 10 V 10 Z 10
X 20 Y 15

c What is the earliest start time after the commencement of the project that task X can commence?
d What is the latest start time after the commencement of the project that task X can
commence without altering the minimum completion from a?
e What is the slack time for task X?
f What is the latest time after the commencement of the project that task W can commence
without altering the minimum completion from a?
g What is the slack time for task W?

7 The times for tasks P to Z shown in the network below are in minutes.
T8
P5 U4 Y4
Q5 V5
R5
Z5
W5
S6
X6
a Determine the minimum completion time and critical path(s) for the above project network.
b How much slack time do activities Q and R have?
8 The times for the tasks shown in the project network below are in days.
C 20 H6
G9
A3
J 10
D 10 E8
K8
B5
F 17

c What is the maximum number of days the
commencement of task F could be delayed without
altering the minimum completion time?
d What float time does activity E have?
motdam/moc.kcotsrettuhS
e What float time does activity H have?

9 The project network below is for the repair of a car at a garage.
90 mins
Order and obtain Test drive
Assess repairs new parts 45 mins
needed
25 mins
Fit new
parts
Remove 30 mins 10 mins
old parts Prepare account
25 mins

b The owner of the car delivers it to the garage at 9 o’clock one morning. Determine the
earliest time it will be ready, assuming the time for each task is as in the network.
c The new parts are in fact all in stock in the garage and do not need to be ordered from
elsewhere. This task then takes just 5 minutes rather than the 90 allowed in the network.
What is the earliest completion time now, given that work commences at 9 a.m.?
10 A cook prepares a meal consisting of casserole, dumplings, fruit salad and chilled wine according
to the project network shown below.
Prepare meat and veg Cook meat and veg
20 mins 75 mins
Prepare dumplings Cook dumplings

5 mins 25 mins
10 mins Prepare
Heat oven fruit salad Set table
Wine in the fridge for chilling 20 mins 5 mins
At least 1 hour
a If the meal has to be ready for 6 p.m. what is the latest time that the first activity can commence?
b What is the latest time that the warm wine can be placed in the fridge for chilling?
c With the wine in the fridge and just as she puts the meat and veg in the oven to cook and is
about to start preparing the dumplings, the cook’s phone rings. What is the maximum time
she can spend on the phone without delaying the completion time? (Assume that she cannot
prepare the dumplings and be on the phone at the same time.)
ONICNAG/moc.kcotSi

Constructing a project network
The previous exercise involved questions being asked about given project networks. In some cases we
may not be given the project network but instead be given sufficient information to allow the network
to be constructed. Examples 2 and 3 are of this type.
EXAMPLE 2
Construct a network for a project consisting of activities A to H listed below.
Activity Time Order

A 7 hours
• Activities A and B start together.
B 9 hours
C 8 hours
• Activities C and D can both commence when A is complete.
D 3 hours
E 7 hours
• Activities F and E can both commence provided both B and D have finished.
F 4 hours
G 5 hours • Activity G can commence provided C and F are both complete.
H 3 hours • Activity H can commence provided E and G are both complete.
Solution A
Activities A and B start together:
B
C
Activities C and D can both commence
A
when A is complete:
D
B
C
Activities F and E can both commence
A F
provided both B and D have finished:
D
B E
C
Activity G can commence provided C G
A F
and F are both complete:
D
B E
C
Activity H can commence provided E G H
and G are both complete: A F
D
B E
C8
Complete the project network by adding G5 H3
the final circle, the arrows and the A7 F4
D3
activity times:
B9 E7

EXAMPLE 3 Task Time Immediate predecessors
A 7 days –
Construct a project network for a B 4 days A
project that consists of the tasks shown C 10 days B
on the right.
D 3 days B
E 4 days B
F 6 days A
G 3 days F
H 7 days F
I 5 days E, G
J 4 days D, H, I
Solution
Start the network with those tasks that A
have no predecessors, in this case task A:
B
Tasks B and F have A as immediate A
predecessor:
F
C
With B in place tasks C, D and E follow: D
B
A E
C
Tasks G and H can be placed: D
B
A E
G
F
H
C
The remaining tasks can be placed: D
B
A E I J
G
F
H
C10
And the network can be completed: D3
B4
A7 E4 I5 J4
G3
F6
H7

Exercise 7B
1 a Construct a project network
Task Scheduled Immediate
for a project consisting of time predecessors
tasks P to X.
P 7 weeks –
b Determine the minimum Q 5 weeks –
completion time and the
R 3 weeks P
critical path.
S 4 weeks P
c How many weeks over the
scheduled time can task U T 6 weeks –
take without altering the U 9 weeks T
minimum completion time? V 5 weeks S, Q
(Assume all other tasks W 7 weeks R
are completed as planned.) X 2 weeks V, W
2 a Construct a network for Task Time Immediate

a project consisting of the predecessors
tasks shown in the table.
Q 6 days –
b Determine the minimum R 8 days –
S 5 days Q
critical path.
T 4 days Q
c Task V is subject to delay
and in fact takes 12 days U 3 days R
to complete, not 7. What V 7 days R
is the minimum completion W 2 days T, U
time now? X 4 days S, W
Y 3 days V, Z
Z 5 days S, W
3 a Construct a project network Task Time Immediate

for a project that consists of predecessors
the tasks shown in the table. A 16 minutes –
b Determine the minimum B 21 minutes A
C 28 minutes –
critical path.
D 25 minutes B, C
c By how many minutes can
activity E be delayed without E 5 minutes B, C
altering the minimum F 25 minutes E
completion time? G 16 minutes D, H
H 17 minutes E

4 a Construct a project network Activity Time Order
for a project consisting of
A 10 days • Activities C and D start together.
activities A to G.
B 21 days • Activities E and F must be after
b Determine the minimum activity C.
completion time and the C 16 days
• Activities A and B need activity
critical path. D 35 days E to be finished before they can
c Lack of harmony on the E 15 days start.
industrial front threatens F 11 days • Activity G can commence only
to delay task F. If all the after F is complete.
G 17 days
other tasks can be completed
according to the times given,
how many extra days can be
allowed for activity F, over
and above the scheduled
11, without increasing the
minimum completion time?
5 a Construct a project network Activity Time Order

for a project consisting of
A 6 hours • Activities A, D and H start
activities A to I.
together.
B 7 hours
b Determine the minimum • Activity E must be after activity A.
completion time and the C 4 hours
• Activities B, F and I must be
critical path. D 7 hours after D.
c Which of the three tasks E 10 hours • Activity C can commence only
that start together, A, D and F 3 hours after both I and H are complete.
H, could experience delays G 5 hours • Activity G can commence only
causing them to be completed after both E and F are complete.
H 9 hours
in a time greater than that
I 4 hours
stated yet still not alter the
minimum completion
time for the project?
6 Construct a project network for Activity Duration Requirements

a project consisting of tasks P to P 5 minutes • Q, T and W start when P is
Z as listed in the table. completed.
Q 6 minutes
Determine the minimum • S can start when T and Y are
R 10 minutes
completed.
completion time and the S 2 minutes • X can start when W is completed.
critical path.
T 4 minutes • R and Y can start when Q is
U 9 minutes completed.
V 4 minutes • Z can start when both S and X
are completed.
W 3 minutes
• U can start when both S and X
X 10 minutes are completed.
Y 4 minutes • V can start when R and Z are
Z 4 minutes completed.

7 a Construct a project network Activity Duration Order
for a project that consists of
R 16 days • R and T start together.
tasks R to Z as listed in the table.
S 22 days • S, U, V start once both R and
b Determine the minimum T are finished.
completion time and the T 25 days
• W starts when S is finished.
critical path. U 20 days
• X starts when W is finished.
c It is thought that activities V 28 days
• Y starts when U is finished.
U, V and Y may each require W 17 days • Z starts when both V and Y
2 extra days for completion X 15 days are finished.
than is given in the table. Y 13 days
What would be the minimum
Z 18 days
completion time if all three of
these tasks do require this extra
time (and all other tasks have
the duration shown in the table).
8 A company decides to interview three of the many applicants for a job. The three people are
Mr James, Ms Kerry and Ms Maine.
The interview process will involve the following tasks:
Activity Description Time Order

(mins)
A Three applicants to meet informally 30 The order in which these activities
with interview panel for coffee should be completed is given below.
B Explain interview procedure to the 5 • Activity A is the first activity and
applicants as a group is followed by B.
C Interview Mr James 45 • Activities C, D and E all follow B.
D Ms Kerry does written test 30 • Activities F, G and H occur once all of

C, D and E have been completed.
E Ms Maine tours work site 30
• Activities I, J and K occur once all of
F Interview Ms Maine 45 F, G and H have been completed.
G Mr James does written test 30 • Activity L follows I, J and K.
H Ms Kerry tours work site 30 • Activity M follows L.
I Interview Ms Kerry 45
J Ms Maine does written test 30
K Mr James tours work site 30
L Interview panel meet to make 45
decision
M Panel chairman informs applicants 2
of panel’s decision
Display the process as a project network.

If activity A, the informal meeting over coffee, finishes at 9.30 a.m. when does Ms Kerry finish
her interview?

9 a Construct a project network for a project consisting of activities Q to Z.
Activity Time Order

Q 4 hours • Q, S and T can all start
R 4 hours without needing other
activities beforehand.
S 7 hours
• R can commence provided Q
T 9 hours has finished.
U 3 hours • U and V can commence
V 11 hours provided R, S and T have all
finished.
W 14 hours
• W, X and Y can commence
X 7 hours
provided U has finished.
Y 9 hours • Z can commence provided V,
Z 3 hours X and Y have all finished.
b Determine the minimum completion time and the critical path.

c Determine how many hours activity Z could run over its allocated time and still not alter
the minimum completion time if
i all the other tasks take exactly their allocated times,
ii task Y is completed in 7 hours rather than the 9 allowed for.
10 A printing company has the job of printing a book for a client. Once they receive the original
pages from the client they carry out the following activities:
Activity Description Time

(days)
A Obtain necessary materials (paper, card etc.) 3
B Paste up originals to form masters 3
C Make printing plates from masters 1
D Print pages 4
E Print cover 1
F Fold and collate pages 2
G Bind pages 3
H Attach cover 2
I Deliver to client 1
The order in which these activities should be completed is given below.

• Obtaining necessary materials and pasting up can start together.
• Making the plates can only be done once the paste up is complete.
• The printing of the pages and the cover can both commence provided both the materials are
obtained and the plates are made.
• The pages can be folded and collated as soon as they are all printed.
• Page binding can occur following folding and collation.
• The printed covers are attached once page binding is complete.
• When all complete the books are delivered to the client.

a Construct a project network for this job.
b Determine the minimum completion time and list the tasks that form the critical path.
c The covers are about to be printed when the machine that prints them breaks down and
a new part has to be sent from Germany. What is the longest time this machine can be out
of use without altering the minimum completion time?
d The company is considering purchasing a new machine that will fold, collate and bind
the pages in a total of 4 days. By how much would this new machine reduce the minimum
completion time?
11 An airline company is keen to reduce the time its

planes are on the ground at a particular airport
because of the high fees the airport authority
charges. The company is charged for each minute
ydribbm/moc.kcotSi
that it has an aircraft at an airport terminal.
Construct a project network for the ‘on ground’
tasks listed below and determine the minimum
completion time.
Activity Description Time (mins) Preceded by

A Support service vehicles in position 5 –
B Passengers disembark 15 A
C Luggage off 8 A
D Old food trolleys off 5 A
E Refuel 10 A
F Security check of empty aircraft 15 B, C
G New passengers on 25 F
H Luggage on 12 F
I Food trolleys on 5 D
J Remove old waste disposal units 8 A
K Fit new waste disposal units 5 J
Which tasks should they attempt to speed up if they are to reduce the time at the terminal?
12 Though generally avoided in this text, the following situation is one requiring the introduction of
a ‘dummy activity’ of zero weight for a network diagram to be formed. Use this idea to represent
the following as a project network with dummy activity Q, and determine the critical path and
shortest completion time.
• Tasks A (5 hours) and B (4 hours) start together.
• Task C (7 hours) can commence when both A and B are completed.
• Task D (6 hours) can commence when task A is completed.

13 In order to keep the production of a school yearbook on time the organising committee list the
tasks involved and the order in which they must occur.
Task Activity Time Order

(weeks)
A Advertise fact that articles are 10 • Tasks A, B and D start together.
required and receive articles • Task C can commence when task B
B Take individual photographs of 3 finishes.
all year twelve students • Task E can commence when task D
C Check all individual photographs 1 finishes.
and retake as necessary • Task F can commence when task A
D Get personal details such as 3 finishes.
nickname etc. from each year 12 • Task G can commence when task F
finishes.
E Have personal details typed to 1
same style and format • Task H can commence when task G
finishes.
F Select articles to be included from 1
those submitted • Task I can commence when tasks C, E
and H are finished.
G Edit selected articles for grammar 1
• Task J can commence when all other tasks
and acceptability
have been completed.
H Have all selected articles typed to 1
same style and format
I Paste up articles, personal details 3
and photos as ordered pages
J Print yearbook 2
Represent this information as a project network and determine the minimum completion time.
Miscellaneous exercise seven

1 Explain what it means to say that the seasonal index for January is 1.12.
2 A loan for $280 000 is taken out with compound interest charged at a rate of 10% per annum
compounded monthly. If repayments of $2800 are made at intervals of one month after the start
of the loan, how much is owed five years after its commencement?
3 An analysis of the number of cars a particular manufacturer sells in each month, over a number
of years, gives the seasonal index for January as 92% and for February as 88%.
The following year the manufacturer sells 2031 cars in January and 1997 cars in February.
Express the number of cars sold in these two months as seasonally adjusted numbers.

4 Which of the following graphs show a an increasing trend?
b a decreasing trend?
c data suitable for linear modelling?
d data with irregular fluctuations?
e data with seasonal fluctuations?
Graph A Graph B
100 100
80 80
60 60
40 40
20 20
5 10 15 t 5 10 15 t
Graph C Graph D
100 100
80 80
60 60
40 40
20 20
5 10 15 t 5 10 15 t
5 Determine the maximum number of units that can be delivered at the sink, G, from the source, A,
for the network shown below.
B 100 D 150 F
200 200 200

150
300 E 500
A G
100
150 150
C
If the capacity of arc AC could be increased from the 150 units shown in the diagram to 250 units
how would this alter the maximum number of units that can be delivered from A to G?

Base
6 The network on the right shows four proposed off-shore
drilling wells and the land-based processing base. Pipelines
are to be laid connecting each well to the base, either 130
A 110 170 140
directly or via one or more of the other wells. The cost 240
of each line is shown in $1000s. D
270
120 170
Determine the network of pipes required to complete
the desired connections whilst keeping the total cost to 90
a minimum, and find this minimum cost. B 160
7 Following the large number of fires that have occurred in a particular state forest over recent
years it is decided that five lookout positions need to be established in the area. A system of
roads is to be developed so that each of the lookout positions is connected to every other lookout
position, either directly or indirectly, by road. The road distances between each pair of lookout
positions would be as given in the table below, in kilometres.
Lookout 1 Lookout 2 Lookout 3 Lookout 4 Lookout 5

Lookout 1 – 5.0 10.5 11.2 6.0
Lookout 2 5.0 – 7.0 10.0 9.0
Lookout 3 10.5 7.0 – 5.5 10.8
Lookout 4 11.2 10.0 5.5 – 7.8
Lookout 5 6.0 9.0 10.8 7.8 –
Which pairs of lookout positions should have direct road connections constructed if the total
length of construction is to be kept to a minimum?
8 The network below shows a system of roads linking town A

to town Z.
Points B, C, D, E and F are road junctions.
C 5 E
4 6
8 4
14
A Z
6 F
6
6 9
7
B 4 D
BEWTSEB/moc.kcotsrettuhS
The number on each road gives the maximum number of

vehicles (in hundreds) that can travel along that section of
road, in the direction indicated, per hour.
Determine the greatest number of vehicles that could arrive
at Z per hour, having come from A.

8.
Assignment
problems
• Assignment problems
• Allocating for a maximum
• Four machines, four tasks
• An algorithm for solving allocation problems
• Suppose we want a maximum
• Suppose the number of people ≠ the number
of tasks
• Miscellaneous exercise eight
Situation
In swimming competitions, a 4 × 100 metre medley relay involves four different swimmers, each
swimming 100 metres and with each doing a different stroke.
The first swimmer swims 100 metres backstroke,
the second swimmer swims 100 metres breaststroke,
the third swimmer swims 100 metres butterfly,
and the fourth swimmer swims 100 metres freestyle.
A swimming club has four swimmers who will form the team for the men’s 4 × 100 metres medley
relay team. They are Alex, Ben, Chris and Devon. The season personal best times for each of these
swimmers, in each of the strokes, over 100 metres are as given in the table below.
(Times are given as minutes.seconds.hundredths-of-a-second. Thus 1.08.72 means 1 minute, 8 seconds
and 72 hundredths of a second, i.e. 68.72 seconds all together.)
Backstroke Breaststroke Butterfly Freestyle

Alex 1.08.72 1.20.17 56.28 58.17
Ben 1.01.24 1.09.21 1.10.21 53.05
Chris 1.03.21 1.17.16 1.03.27 55.21
Devon 1.04.17 1.08.71 1.17.24 54.37
Which swimmer would you have swimming each leg of the relay to give the fastest possible time for
the team, based on the season personal best times?
Suppose instead that Alex’s season personal best time for the butterfly is 1.08.17. I.e. the table of season
personal best times would now be as follows:

Alex 1.08.72 1.20.17 1.08.17 58.17
Ben 1.01.24 1.09.21 1.10.21 53.05
Chris 1.03.21 1.17.16 1.03.27 55.21
Devon 1.04.17 1.08.71 1.17.24 54.37
What would be the preferred order now? 1mredcmc/moc.kcotSi
ISBN 9780170395069 8. Assignment problems 141

The situation on the previous page required us to assign each swimmer to a stroke in order to give the
fastest medley relay team, based on season personal best times. This chapter considers assignment
problems of this type, also called allocation problems. The swimming situation required us to
allocate the swimmers in order to give the best total time. Not all problems of this type involve
optimising time. Some may be concerned with optimising cost or distance travelled etc.
Assignment problems
Suppose a company wishes to use two transport companies, A and B, for two deliveries, 1 and 2,
giving one delivery to each transport company. The locations of each pick-up point and delivery
point in relation to each transport company, and the transport company costs, means that for each
job, each of the transport companies quotes a different price. The quoted prices are as follows:
Delivery 1 Delivery 2
Company A $145 $180
Company B $164 $191
Alternatively, this information could be given as a bipartite graph:
A B
$1 64
80 $1
$
19
541
1$
1 2
Can you see which allocation of the two jobs will minimise the total cost (remembering that we must
give one delivery to each company)?
If we write AB to indicate company A doing delivery 1 and company B doing delivery 2, there are only
two allocations that are possible:
AB and BA
AB: Company A doing delivery 1 and company B doing delivery 2.
Total cost = $145 + $191
= $336
BA: Company B doing delivery 1 and company A doing delivery 2.
Total cost = $164 + $180
= $344
Cheapest solution: Company A doing delivery 1 and company B doing delivery 2 for a total cost of $336.
Is this the solution you chose just by considering the table of costs?

Now let us consider a situation involving three companies, A, B and C, and three deliveries, 1, 2 and 3,
again with one delivery to be assigned to each company.
Suppose the costs this time are as follows: A
$170
1
0
$16
90
1 2 3 $1
$15
A $170 $150 $260 0
$160
B $160 $160 $210 B 2
0
$17
C $190 $170 $270 $2
$21 60
0
$270
C 3
Now, writing ABC to mean company A does delivery 1, B does delivery 2 and C does delivery 3, there
are six possible ways the assignments can be organised:
ABC: $170 + $160 + $270 = $600
ACB: $170 + $170 + $210 = $550 ←
BAC: $160 + $150 + $270 = $580

BCA: $160 + $170 + $260 = $590
CAB: $190 + $150 + $210 = $550 ←
CBA: $190 + $160 + $260 = $610

Thus there are two allocations that equally give the minimum cost of $550:
A do delivery 1, B do delivery 3 and C do delivery 2,
or
A do delivery 2, B do delivery 3 and C do delivery 1.
Allocating for a maximum

Some allocation problems may involve situations in which the most desirable, or optimum, situation
involves a maximum rather than a minimum. Consider for example allocating three workers to three
machines. Each machine makes different components and each worker has different competencies on
each machine. However the company wants to allocate each worker to a machine such that the total
number of components produced is maximised. Suppose that the number of components produced
in a day by each worker, A, B and C, on each machine, 1, 2 and 3, is as follows:
Machine 1 Machine 2 Machine 3

Worker A 1020 950 1430
Worker B 760 920 1200
Worker C 950 1000 1450
Which worker should be assigned to each machine to maximise the total number of components
produced by the three workers?
Try to find a solution before turning the page where a solution is presented. Perhaps set up a
spreadsheet showing the various arrangements and their totals.

Writing ABC to mean worker A is on machine 1, B is on machine 2 and C is on machine 3, there are
six possible ways the workers can be assigned to machines:
ABC: 1020 + 920 + 1450 = 3390 ← maximum total.
ACB: 1020 + 1000 + 1200 = 3220
BAC: 760 + 950 + 1450 = 3160
BCA: 760 + 1000 + 1430 = 3190
CAB: 950 + 950 + 1200 = 3100
CBA: 950 + 920 + 1430 = 3300
Hence we should assign worker A to machine 1, worker B to machine 2 and worker C to machine 3 for
a total number of components produced of 3390.
Four machines, four tasks

The next example returns to an optimum situation involving a minimum but now a 4 × 4 table is involved.
EXAMPLE 1
Four machines are to be assigned to carry Task 1 Task 2 Task 3 Task 4

out four tasks, with one machine for each
Machine A 15 18 9 12
task, and the aim of minimising the total
machine time taken to complete the tasks. Machine B 13 21 11 15
Machine C 12 19 8 13
The time each machine is estimated
to complete each task is given on the right, Machine D 14 24 10 11
in hours. Find the allocation of machines
to tasks that will minimise total machine time.
Solution
Writing ABCD to mean machine A is on task 1, machine B is on task 2, machine C is on task 3 and
machine D is on task 4, there are 24 possible ways the machines can be assigned to the tasks:
ABCD: 15 + 21 + 8 + 11 = 55 CABD: 12 + 18 + 11 + 11 = 52
ABDC: 15 + 21 + 10 + 13 = 59 CADB: 12 + 18 + 10 + 15 = 55
ACBD: 15 + 19 + 11 + 11 = 56 CBAD: 12 + 21 + 9 + 11 = 53
ACDB: 15 + 19 + 10 + 15 = 59 CBDA: 12 + 21 + 10 + 12 = 55
ADBC: 15 + 24 + 11 + 13 = 63 CDAB: 12 + 24 + 9 + 15 = 60
ADCB: 15 + 24 + 8 + 15 = 62 CDBA: 12 + 24 + 11 + 12 = 59
BACD: 13 + 18 + 8 + 11 = 50 ← DABC: 14 + 18 + 11 + 13 = 56
BADC: 13 + 18 + 10 + 13 = 54 DACB: 14 + 18 + 8 + 15 = 55
BCAD: 13 + 19 + 9 + 11 = 52 DBAC: 14 + 21 + 9 + 13 = 57
BCDA: 13 + 19 + 10 + 12 = 54 DBCA: 14 + 21 + 8 + 12 = 55
BDAC: 13 + 24 + 9 + 13 = 59 DCAB: 14 + 19 + 9 + 15 = 57
BDCA: 13 + 24 + 8 + 12 = 57 DCBA: 14 + 19 + 11 + 12 = 56
Assign machine A to task 2, B to task 1, C to task 3 and D to task 4. Total time 50 hours.
Are you able to set up a spreadsheet for the above listing that would be adaptable for any other
‘four machine, four task’ situation?

Exercise 8A
The following tables show the cost that trucking firms A, B, C, … charge to deliver loads 1, 2, 3, … .
Assign trucking firms to loads such that in each case each firm is assigned a different load and the total
cost is minimised.
1 Load 1 Load 2
Firm A $150 $120
Firm B $120 $110
2 Load 1 Load 2 Load 3

Firm A $100 $140 $150
Firm B $90 $170 $100
Firm C $110 $140 $130
3 Load 1 Load 2 Load 3 Load 4

Firm A $700 $650 $850 $1200
Firm B $600 $700 $950 $1050
Firm C $750 $600 $750 $1250
Firm D $650 $750 $850 $1150
The following tables show the number of components each operator, A, B, C, … can produce in a day
when operating machines 1, 2, 3, … .
Assign operators to machines such that in each case each operator is assigned to a different machine
and the total number of components produced in a day is maximised.
4 Machine 1 Machine 2
Operator A 80 90
Operator B 85 100
5 Machine 1 Machine 2 Machine 3

Operator A 210 200 170
Operator B 230 210 190
Operator C 200 180 180
6 Machine 1 Machine 2 Machine 3 Machine 4

Operator A 60 90 105 80
Operator B 80 90 110 85
Operator C 65 95 110 90
Operator D 75 85 115 75

WS
An algorithm for solving allocation problems
Whilst the process of listing all possibilities to determine the optimum arrangement is a valid process,
Allocation and the
Hungarian algorithm there is an algorithm we can use to determine the optimum allocation, without having to list all
possibilities. The steps of this algorithm are shown below.
Consider the situation of assigning four deliveries to four courier firms with each firm taking one
delivery and with the allocation aiming to minimise the total cost.
Suppose the costs charged by each firm are as follows
Delivery 1 Delivery 2 Delivery 3 Delivery 4

Courier firm A $250 $100 $140 $120
Courier firm B $220 $110 $150 $110
Courier firm C $220 $90 $160 $130
Courier firm D $260 $140 $170 $120
250 100 140 120

Step 1: Identify the cost matrix: 220 110 150 110
220 90 160 130
260 140 170 120
Step 2: Subtract the smallest number in each row from every number in that row.
Hence we subtract 100 from each element of row 1,
110 from each element of row 2,
90 from each element of row 3,
and 120 from each element of row 4.
150 0 40 20
This gives the matrix: 110 0 40 0
130 0 70 40
140 20 50 0
Step 3: Subtract the smallest number in each column from every entry in that column.
Hence we subtract 110 from each element of column 1,
0 from each element of column 2,
40 from each element of column 3,
and 0 from each element of column 4.
40 0 0 20
This gives the matrix: 0 0 0 0
20 0 30 40
30 20 10 0
Step 4: Now draw as few horizontal and/or vertical straight lines 40 0 0 20

as possible through all of the zeros. This question involves 0 0 0 0
a 4 × 4 matrix so if the number of such lines equals 4, a 20 0 30 40
solution can be found from this matrix, as is the case here. 30 20 10 0

Step 5: We now need to choose one zero in each column and one zero in each row.
(Hint: First consider any rows and columns that only have one zero entry in them as that
entry must be chosen.)
40 0 0 20
0 0 0 0
20 0 30 40
30 20 10 0
The optimum situation is to assign courier firm A to delivery 3,

courier firm B to delivery 1,
courier firm C to delivery 2,
and courier firm D to delivery 4.
This would have a total cost of $140 + $220 + $90 + $120 = $570
Note: • Step 4, the drawing of the lines could be done mentally.
• If in step 4 the zeros are crossed through using less lines than there are rows (or columns)
in the matrix, further adjustment of the matrix is necessary, as will be explained soon.
• The listing of all 24 possible allocations of courier firms to deliveries is shown below.
The reader should confirm that the optimum solution given by the algorithm is the
same as obtained from the list of all possibilities.
ABCD: $250 + $110 + $160 + $120 = $640
ABDC: $250 + $110 + $170 + $130 = $660
ACBD: $250 + $90 + $150 + $120 = $610
ACDB: $250 + $90 + $170 + $110 = $620
ADBC: $250 + $140 + $150 + $130 = $670
ADCB: $250 + $140 + $160 + $110 = $660
BACD: $220 + $100 + $160 + $120 = $600
BADC: $220 + $100 + $170 + $130 = $620
BCAD: $220 + $90 + $140 + $120 = $570
BCDA: $220 + $90 + $170 + $120 = $600
BDAC: $220 + $140 + $140 + $130 = $630
BDCA: $220 + $140 + $160 + $120 = $640
CABD: $220 + $100 + $150 + $120 = $590
CADB: $220 + $100 + $170 + $110 = $600
CBAD: $220 + $110 + $140 + $120 = $590
CBDA: $220 + $110 + $170 + $120 = $620
CDAB: $220 + $140 + $140 + $110 = $610
CDBA: $220 + $140 + $150 + $120 = $630
DABC: $260 + $100 + $150 + $130 = $640
DACB: $260 + $100 + $160 + $110 = $630
DBAC: $260 + $110 + $140 + $130 = $640
DBCA: $260 + $110 + $160 + $120 = $650
DCAB: $260 + $90 + $140 + $110 = $600
DCBA: $260 + $90 + $150 + $120 = $620

Question 2 of Exercise 8A involved the following information:
Load 1 Load 2 Load 3

Firm A $100 $140 $150
Firm B $90 $170 $100
Firm C $110 $140 $130
To minimise total cost the allocation was: firm A to do load 1,

firm B to do load 3,
firm C to do load 2.
The total cost would then be $340.
Follow the steps of the algorithm as shown below to see that it gives the same solution:
The cost matrix: 100 140 150

90 170 100
110 140 130
Subtract the smallest number in each row from every 0 40 50

number in that row. 0 80 10
0 30 20
Subtract the smallest number in each column from every 0 10 40

entry in that column. 0 50 0
0 0 10
Draw as few horizontal and/or vertical straight lines 0 10 40

as possible through all of the zeros. If the number 0 50 0
of lines equals the number of rows (or columns) of the 0 0 10
matrix then a solution can be found from this matrix.
Determine the solution and relate identified locations 0 10 40

in matrix back to the original. 0 50 0
0 0 10
I.e., as before, firm A to do load 1 ($100), firm B to do load 3 ($100) and firm C to do load 2 ($140).
Total cost $340.

If drawing the horizontal and/or vertical lines on the matrix to cover the zeros can be done using fewer
lines than there are rows (or columns) in the matrix we need to adjust the matrix so that a solution can
be found.
For example, consider:
1 2 3
A 8 5 17
B 11 7 18
C 9 8 15
The cost matrix: 8 5 17

11 7 18
9 8 15
Subtract the smallest number in each row from 3 0 12

every number in that row. 4 0 11
1 0 7
Subtract the smallest number in each column from 2 0 5

every entry in that column. 3 0 4
0 0 0
Draw as few horizontal and/or vertical straight lines 2 0 5

as possible through all of the zeros. The number of 3 0 4
lines is less than the number of rows (or columns) 0 0 0
the matrix has. Further adjustments needed.
For the numbers that do not have a line drawn 0 0 3
through them (2, 5, 3, 4), take the smallest of 1 0 2
these numbers (2) from each of these numbers 0 2 0
and also add this smallest number to any location
where two of the straight lines intersect.
The zeros now require three lines so a solution 0 0 3
is possible (if not repeat the previous two steps). 1 0 2
Determine the solution and relate the identified 0 2 0
locations in matrix back to the original.
A should be assigned to task 1. ‘Cost’ is 8 units.
B should be assigned to task 2. ‘Cost’ is 7 units.
C should be assigned to task 3. ‘Cost’ is 15 units.
Total ‘cost’ is 30 units.

The algorithm we are using here is called The Hungarian Algorithm, so named by Harold Kuhn
in 1955, because the method was based on previous work by two Hungarian mathematicians, König
and Egerváry.
Three examples of the algorithm are shown below. In each case the initial table shows the ‘cost’ of
allocating each of A, B, C, … to tasks 1, 2, 3, … This ‘cost’ could be dollars, or hours, or distance
travelled etc, and the task is to minimise the total cost. Check through each example to confirm that
you too could obtain the solutions in this way.
1 2 3 4 1 2 3 4 1 2 3
A 8 7 15 6 A 60 50 120 70 A 80 60 90
B 10 6 15 7 B 50 60 100 80 B 100 70 110
C 11 10 12 5 C 55 80 140 80 C 90 100 100
D 9 7 13 8 D 80 80 110 70
↓
↓ ↓
80 60 90
8 7 15 6 60 50 120 70 100 70 110
10 6 15 7 50 60 100 80 90 100 100
11 10 12 5 55 80 140 80 ↓
9 7 13 8 80 80 110 70
20 0 30
↓ ↓
30 0 40
2 1 9 0 10 0 70 20 0 10 10
4 0 9 1 0 10 50 30 ↓
6 5 7 0 0 25 85 25
2 0 6 1 10 10 40 0 20 0 20
30 0 30
↓ ↓
0 10 0
0 1 3 0 10 0 30 20 ↓
2 0 3 1 0 10 10 30
4 5 1 0 0 25 45 25 0 0 0
0 0 0 1 10 10 0 0 10 0 10
0 30 0
↓ ↓
↓
0 1 3 0 10 0 20 10
Multiple solutions.
2 0 3 1 0 10 0 20
Assign B to 2
4 5 1 0 0 25 35 15
Cost = 70
0 0 0 1 20 20 0 0 Then either
Assign A to 1. Cost = 8 Assign A to 2. Cost = 50 A to 1 (cost 80) and
Assign B to 2. Cost = 6 Assign B to 3. Cost = 100 C to 3 (cost 100).
Assign C to 4. Cost = 5 Assign C to 1. Cost = 55 Or
Assign D to 3. Cost = 13 Assign D to 4. Cost = 70 A to 3 (cost 90) and
C to 1 (cost 90).
Total cost = 32 Total cost = 275
Either way:
Total cost = 250

Note: The Hungarian algorithm can be used when we are assigning one ‘person per task’ and one
‘task per person’. It relies on the fact, not proven here, that adding the same number to every
element in a row or column of the cost matrix (or subtracting the same number from every
element in a row or column of the cost matrix) does not change the ‘location’ of the optimum
arrangement. Once we have this location we can return to the original matrix to obtain the
cost details.
Exercise 8B
For each of questions 1 to 4 use the Hungarian algorithm to allocate the companies A, B, C, … to the
tasks 1, 2, 3, … such that each company is allocated a different task and the total ‘cost’ is minimised,
where the units of ‘cost’ of each item being assigned to each task is as given by the numbers in the table
or bipartite graph. As well as stating the assignments of company to task, also state the total cost.
1 Task 1 Task 2
Company A 95 72
Company B 105 86
2 Task 1 Task 2 Task 3

Company A 70 110 120
Company B 50 140 150
Company C 100 170 140
3 A B C D
10 15 10 14 9 12 14 15 8 15 10 13 9 11 10 14
1 2 3 4
4 Task 1 Task 2 Task 3 Task 4 Task 5 Task 6

Company A 20 30 20 30 20 30
Company B 24 20 12 35 10 28
Company C 30 25 20 55 21 38
Company D 25 44 21 36 20 30
Company E 20 32 20 40 22 35
Company F 40 50 20 48 25 40

For each of questions 5 to 9 use the Hungarian algorithm to allocate the companies A, B, C, … to the
tasks 1, 2, 3, … such that each company is allocated a different task and the total ‘benefit’ is maximised,
where the units of ‘benefit’ of each company being assigned to each task is as given by the numbers in
the table or bipartite graph. As well as stating the assignment of company to tasks also state the benefit
of each assignment.
5 Task 1 Task 2
Company A 250 120
Company B 240 90
6 Task 1 Task 2 Task 3

Company A 29 29 31
Company B 26 24 28
Company C 22 33 22
7 Task 1 Task 2 Task 3 Task 4

Company A 120 240 120 560
Company B 150 250 140 620
Company C 110 220 100 610
Company D 160 210 90 540
8 A 44
1
9 A 16 1
40 26
43
33
31
29
51
34 B 30 2
B 2
42 29
43
35
38 33
20
C 3 C 32 3
42
44
20
45
23
D 36 4

10 A company has four regional ‘sales reps’ with one situated in Adelaide, one in Brisbane, one in
Hobart and one in Perth. The company plans to open offices in Broome, Darwin, Melbourne
and Sydney and, at least initially, wants each of the existing sales reps to oversee the new reps in
these new offices, with one existing rep per new office. This overseeing role is likely to involve
a fair bit of travel so the company decides to attach each existing rep to the new office in a way
that minimises the total cost of a return airfare for each existing rep to visit the new office they
are overseeing.
The return airfares between the locations are as follows:
Broome Darwin Melbourne Sydney

Adelaide $1080 $780 $240 $380
Brisbane $1460 $570 $340 $270
Hobart $1750 $1120 $290 $550
Perth $720 $680 $440 $580
a Just by looking at the table try to write down what you think the optimal allocation of
existing reps to new offices would be.
b Clearly showing your use of the Hungarian algorithm determine the optimal allocation.
11 a Allocate four delivery jobs to four courier companies with each company getting one of the
jobs and total costs being kept to a minimum, and state this minimum, given that the cost
of each job with each company is as shown below.
Job 1 Job 2 Job 3 Job 4

Quick Courier Co. $290 $70 $290 $150
Speedy Courier Co. $320 $70 $260 $160
Deliverit Couriers $460 $80 $320 $180
Competitive Couriers $280 $60 $250 $140
b Suppose the job description for job 4 changed, causing every one of the four courier
companies to increase their price for job 4 by $50, all other jobs remaining the same as
in the above table. Would this alter the optimum allocation of company to job? If it would,
state the new allocation.
c Suppose instead that it was job 2 that had its job description changed and this caused all
four companies to multiply their price for that job by 4, all other jobs remaining as in the
table. Would this alter the optimum allocation of company to job? If it would, state the
new allocation.
d Suppose instead that just one of the companies is to be used for all four jobs. Which
company would this be if again the aim is to minimise total cost?

12 A real estate company is allocating areas
to each of its five salespeople, Jack, Jim,
Judy, Nicci and Ti. Each salesperson is to
have their own area, with the five areas
labelled 1, 2, 3, 4 and 5.
An analysis of the areas, the selling
grebmoolB/segamI ytteG
history of the representatives and their
existing profile in the areas suggest that
the number of houses each representative
should sell in a year, were they assigned
to a particular area, would be as follows.
Area 1 Area 2 Area 3 Area 4 Area 5

Jack 10 9 8 7 6
Jim 5 7 7 9 7
Judy 9 7 9 10 8
Nicci 8 6 10 10 10
Ti 11 12 8 13 10
Allocate the representatives to the areas in such a way that will maximise the total number of
houses the five representatives should sell in a year.
What will this maximum total be?
13 A mining company employs four new engineers, Ashok, Daniel, Kirsten and Shani, and wishes
to locate one at each of its four major mining locations. Travel and relocation expenses are
available to all four and the company decides to send the engineers to the location that minimises
the total travel and relocation costs. These costs are estimated to be as follows.
Ashok Daniel Kirsten Shani

$800 00 00
0 $90 $14 0
000 21$
000 21$
000 41$
$1
$9 00
00 00
0 00
70
$1
0008$
00 $1
00 $13 0
000 000
$1 00
0
0
$1 000
$70
00
$15
Location Location Location Location
1 2 3 4
Allocate each engineer to the appropriate mining location.

14 A construction company has one mobile crane available from each of its five equipment storage
bases. It needs to send these five mobile cranes to five job locations, with one crane per location.
The distances involved for each crane to get to each location are as shown in the following table,
in kilometres.
Location 1 Location 2 Location 3 Location 4 Location 5

Crane A 40 90 30 70 60
Crane B 30 70 40 60 30
Crane C 40 30 70 40 30
Crane D 50 80 40 50 90
Crane E 50 60 40 40 70
Allocate each crane to a location such that the total distance travelled is minimised.
15 A car making company has five factories, A, B, C, D and E each producing all of the five major
models of car that the company produces:
The Arcain The Bijou The Charger The Devine The Exeon
512g/moc.kcotsrettuhS
The company wishes to reorganise the factories and have one factory only producing the Arcain,
another only producing the Bijou and so on.
Estimates suggest that the numbers each factory could produce of each model in a year, if they
were dedicated to only producing the one model, would be as follows.
Arcain Bijou Charger Devine Exeon

Factory A 230 000 250 000 200 000 190 000 180 000
Factory B 150 000 150 000 130 000 120 000 120 000
Factory C 320 000 330 000 300 000 280 000 290 000
Factory D 180 000 170 000 120 000 150 000 140 000
Factory E 240 000 200 000 200 000 180 000 180 000
The Bijou is the best-selling model so the company knows it wants to locate the construction
of that model at its biggest factory, Factory C.
Allocate a factory to each model so each factory produces only one model, all five models
are produced, the Bijou is at factory C, and the total number of models produced in a year
is maximised.

Suppose the number of people ≠ the number of tasks
In all of the assignment situations encountered so far in this chapter, the number of items we have
been assigning (be they swimmers, trucking companies, machine operators, etc) has always been the
same as the number of things we have assigned these items to (be they swimming legs in a relay, loads
to be delivered, machines, etc.). Such assignment problems are sometimes referred to as being balanced.
The cost matrix will be a square matrix. Suppose the number of items we are assigning does not match
the number of items we are assigning them to? In such cases we assign a dummy row, or dummy column,
as appropriate, with all of its entries equal to zero. In this way we create a square matrix and the
Hungarian algorithm can be used. The next two examples illustrate this technique.
EXAMPLE 2
The director of a company decides that he wants a representative from the company at each of the
three international trade conferences that are coming up. The director wishes to send a different
manager to each, chosen from the four managers the company has based at its various branches,
making the selection on the basis of the arrangement that keeps travel costs to a minimum. The
associated travel costs are as follows:
Conference 1 Conference 2 Conference 3

Manager A $1040 $780 $2140
Manager B $1150 $560 $1750
Manager C $1780 $1250 $1350
Manager D $1560 $790 $1215
What should the allocation of managers be?
Solution
Inserting a dummy column with all entries equal 1040 780 2140 0
to zero gives the square matrix on the right. 1150 560 1750 0
With a zero in each row subtracting the lowest 1780 1250 1350 0
number in each row from every entry in that row 1560 790 1215 0
will leave the matrix unchanged.
Subtract the smallest number in each column 0 220 925 0
from every number in that column. 110 0 535 0
740 690 135 0
520 230 0 0
Four horizontal and/or vertical lines are needed 0 220 925 0

to cover all of the zeros so a solution can be found 110 0 535 0
from this matrix. 740 690 135 0
520 230 0 0

Determine the solution: 0 220 925 0
110 0 535 0
740 690 135 0
520 230 0 0
Relate identified locations in the matrix 1 2 3

back to original, remembering that the final A 1040 780 2140 0
column is for a non-existent conference. B 1150 560 1750 0
C 1780 1250 1350 0
D 1560 790 1215 0
Hence send manager A to conference 1. (for a cost of $1040)

Send manager B to conference 2. (for a cost of $560)
Send manager D to conference 3. (for a cost of $1215)
(Total cost: $2815)
Instead of making a dummy column with all entries equal to zero, try the previous example with
the dummy column having all of its entries equal to the largest number in the matrix (i.e. 2140).
Does that give the same solution?
EXAMPLE 3
An international company has three locations across the world, each with its own director. The
three directors decide they should each make a fact-finding trip to one of four other countries to
explore export opportunities. They choose four countries and decide that between them they will
visit three of these four with each director visiting one country. The associated travel costs for each
director to visit each country are as follows:
Country 1 Country 2 Country 3 Country 4

Director A $10100 $13300 $11500 $7200
Director B $9300 $8200 $7900 $6300
Director C $12700 $11700 $14200 $9500
Allocate each director to one of the countries so that each is attending a different country from the
other two and total travel costs are kept to a minimum.
Solution
Inserting a dummy row with all entries equal 10100 13300 11500 7200
to zero gives the square matrix on the right. 9300 8200 7900 6300
12700 11700 14200 9500
0 0 0 0

Subtracting the smallest number in each row 2900 6100 4300 0
from every number in that row gives the matrix 3000 1900 1600 0
on the right. 3200 2200 4700 0
With a zero in each column, subtracting 0 0 0 0
the smallest number in each column will not
change the matrix.
Drawing horizontal and vertical lines indicates 2900 6100 4300 0
further adjustment needed. 3000 1900 1600 0
3200 2200 4700 0
0 0 0 0
Adjust: 1300 4500 2700 0

1400 300 0 0
1600 600 3100 0
0 0 0 1600
Drawing lines indicates yet further 1300 4500 2700 0

adjustment needed. 1400 300 0 0
1600 600 3100 0
0 0 0 1600
Adjust: 700 3900 2100 0

1400 300 0 600
1000 0 2500 0
0 0 0 2200
Drawing lines indicates a solution can now 700 3900 2100 0

be determined. 1400 300 0 600
1000 0 2500 0
0 0 0 2200
Determine the solution and relate identified 700 3900 2100 0

locations in matrix back to original, remembering 1400 300 0 600
that the bottom row is for a non-existent director. 1000 0 2500 0
0 0 0 2200
Director A should visit country 4. (Cost: $7200)

Director B should visit country 3. (Cost: $7900)
Director C should visit country 2. (Cost: $11 700)
(Total cost: $26 800)

Exercise 8C
1 Three people, Alex, Ben and Theo, are each to be assigned a task from tasks 1, 2, 3 and 4, with
each person having just one task, i.e. just three of the tasks will be assigned. The time each person
takes to do each task, in minutes, is as shown below. The assignment of each person to a task is to
be done in such a way that the total of the three times is to minimised.
Clearly showing your use of the Hungarian algorithm, assign each person to a task according to
the above requirements.
Alex Ben Theo
24 23 15 20 18 25 20 17 21 28 18 24
Task 1 Task 2 Task 3 Task 4
Under this minimal total time arrangement, if all three people start their assigned task at the same
time, how many minutes later will all three have finished (assuming any early finishers do not assist
others with their task)? Is there a different allocation arrangement that would reduce this ‘time to
finish three jobs when working simultaneously’?
2 An equipment hire company is asked to supply

three industrial diggers, one to a worksite in
Denton, one to a worksite in Abbysix and one
to a worksite in Crampon.
kuylyryK rymydoloV/moc.kcotsrettuhS
The company has four such diggers available,
one at the Horton storage facility, one at the
Peel storage facility, one at the Washton storage
facility and the fourth at the Wimsley storage
facility. The distances, in kilometres, that each
of these storage facilities is from each of the
worksites is given in the following table:
Denton Abbysix Crampon

Horton storage facility 27 km 25 km 25 km
Peel storage facility 58 km 22 km 18 km
Washton storage facility 34 km 36 km 51 km
Wimsley storage facility 31 km 55 km 54 km
Clearly showing your use of the Hungarian algorithm, assign three of the diggers, one to each
worksite, such that the total distance to be travelled by the diggers to the worksites is minimised,
and state this minimum distance.

3 A taxi company has four available taxis at Bill Zarani Tenielle
various locations. The company receives
Taxi A 9 7 8
three calls from separate individuals, Bill,
Zarani and Tenielle, each requesting a taxi. Taxi B 11 7 10
The estimated times each available taxi will Taxi C 10 6 8

take to get to each of the people requesting Taxi D 8 8 9
a taxi are as shown in the table.
Assign a taxi to each person such that the total time taken by the three taxis used is the minimum
possible, and state the time each person has to wait.
LmmoT/moc.kcotSi
4 Four people from A, B, C, D and E are to be assigned jobs 1, 2, 3 and 4 with a different person for
each job and one of the five people ‘missing out’.
The profit that will be made from each person doing each job is as shown in the table below.

A $75 $175 $150 $120
B $90 $220 $150 $130
C $120 $240 $170 $160
D $100 $200 $160 $140
E $90 $180 $150 $150
Supposing you are going to use the Hungarian algorithm to assign people to jobs according to
the above requirements and in a way that maximises the profit to the company.
Which are you going to do first:
• put in the dummy column and then subtract each number from the largest number, or
• subtract each number from the largest number and then put in the dummy column, or
• doesn’t it matter?
Investigate, and find the solution to the assignment problem too, stating which person gets which
job and the maximum profit.

Challenge
Time-and-motion experts are looking at the assignment of ten workers to ten tasks, one worker
per task. The experts first measure the time taken by each of the workers to complete each task.
They find that the varied experience and skill levels of the workers mean that not all workers
complete the same task in the same time. The various times taken by each worker on each task,
in minutes, is as given below.
Tasks
1 2 3 4 5 6 7 8 9 10
Worker A 17 21 9 28 15 19 33 45 15 14
Worker B 15 19 10 25 17 15 28 37 28 19
Worker C 22 22 13 21 17 22 47 46 17 17
Worker D 16 22 10 23 16 17 29 39 22 21
Worker E 18 20 9 29 21 18 32 40 23 14
Worker F 18 21 11 30 17 24 33 38 19 15
Worker G 17 23 12 22 16 18 45 40 24 21
Worker H 25 20 15 23 15 28 29 42 25 22
Worker I 16 20 10 25 14 17 32 55 20 18
Worker J 17 24 12 28 17 21 28 36 20 15
At present the workers are assigned as follows:

Worker A to task 1 (17 mins) Worker B to task 2 (19 mins)
Worker C to task 3 (13 mins) Worker D to task 4 (23 mins)
Worker E to task 5 (21 mins) Worker F to task 6 (24 mins)
Worker G to task 7 (45 mins) Worker H to task 8 (42 mins)
Worker I to task 9 (20 mins) Worker J to task 10 (15 mins)
Summing these times gives a total of 239 minutes.
The experts want to find the allocation of worker to task that would minimise this total.
1. Solve the above allocation problem yourself using the Hungarian algorithm.
2. See if there is an online Hungarian algorithm solver and, if so, see if it gives the same solution
to the above problem as you obtained.
hcizelyW nreojB/moc.kcotsrettuhS

Miscellaneous exercise eight

Annual 10%
Six monthly 10%
Quarterly 10%
Monthly 10%
Weekly 10%
Daily 10%
2 To the nearest dollar, how much should be invested in an annuity paying 10% annual interest,
compounded annually, to provide a regular annual income of $65 000 for exactly 20 years?
a using a recursive formula with an initial ‘guess’ of $560 000, and
3 Suppose instead that the annual interest rate in the previous question had been 7.5% rather than
10%. Initially make a guess at how much would need to be invested to provide the same annual
income, i.e. $65 000 for exactly 20 years, and then use recursion to determine the correct answer,
to the nearest dollar.
4 A loan for $8000 is taken out with compound interest charged at a rate of 8% per annum
compounded monthly. If repayments of $250 are made at intervals of one month after the
start of the loan how much is owed immediately after the fifteenth repayment has been made?
5 The times for tasks A,

B, C etc., shown in the
A5 G5
project network on the
right are in hours. E7 L3
B7 H4
(The letter I has been
deliberately omitted.)
Determine the C3 J7
minimum completion F8 M6
time and critical path D4 K2

for the network.

6 The network below shows the connections in a telecommunications system. The vertices represent
the switching stations, capable of sending messages from one station to another. The numbers on
the edges give the number of messages that can be sent from one station to another in an hour.
a Find the greatest number B 400 E
of messages that can go 500
250 F
from A to G in an hour. 100 100
450 600
b Find the greatest number A
G
C 100
of messages that can go 100
from A to D in an hour. 250
300
D
7 The network below shows a system of pathways that allow units to flow from a source A to a sink X.
The units that can flow along each pathway are as indicated by the numbers next to each pathway.
Find the maximum number of units that can flow from A to X (and check your answer by finding
a cut equal to your maximum flow).
C 100 G
250 300 500
200
300
A X
400 600 E
400 D 100 600

750
B 400 F
If the capacity of pathway EX were to be increased by 300 units, how would this alter the
maximum flow from A to X?
8 Four swimmers, Julie, Micah, Fran and Marta are chosen to swim the medley relay for their
school at the interschool swimming carnival. Each of the girls will swim one ‘leg’ of the relay.
One will swim 50 metres backstroke, one will swim 50 metres breaststroke, one will swim
50 metres butterfly and one will swim 50 metres freestyle.
To decide which girl swims each leg, the coach times each of them over 50 metres of each style.
The times recorded, in seconds, were as follows:

Julie 48.6 56.8 43.6 38.4
Micah 56.8 61.3 55.8 42.4
Fran 54.2 52.4 62.7 64.7
Marta 72.4 48.2 65.3 62.3
Use these times to suggest the most appropriate allocation of girls to strokes and suggest the total
time for such a team.

9 The following table gives the numbers of live cattle exported from Australia each year from 1988
to 2013. (Numbers are given to the nearest one hundred).
Year 1988 1989 1990 1991 1992 1993

Number of cattle 81 500 83 300 90 200 94 700 132 300 190 400
Year 1994 1995 1996 1997 1998 1999

Number of cattle 290 900 480 800 724 100 912 900 597 000 833 700
Year 2000 2001 2002 2003 2004 2005

Number of cattle 887 000 797 900 955 100 689 400 570 800 531 200
Year 2006 2007 2008 2009 2010 2011

Number of cattle 609 200 668 900 815 600 921 300 821 600 621 900
Year 2012 2013

Number of cattle 516 500 726 000
Source of data: Australian Bureau of Statistics.
With the assistance of a computer spreadsheet draw

both the raw data and the five-point moving averages
on the same time series line graph.
Write some sentences describing any trends in the
okjodaR okjleZ/moc.kcotsrettuhS
numbers of live cattle exported from Australia per
year over the period of time from 1988 to 2013.
10 Five police vehicles, A, B, C, D and E have to be sent to five locations, 1, 2, 3, 4 and 5, with one
police vehicle per location. The estimated time, in minutes, for each police vehicle to reach each
location, from where they are now, is given in the following bipartite graph.
A B C D E
7 12 7 11 8 6 9 12 12 12 6 12 9 10 11 6 9 9 11 9 7 11 9 10 11
1 2 3 4 5
Allocate the police vehicles to the locations in such a way that the sum of the five travel times is
minimised, and state the time taken for each vehicle to reach its allocated location.

11 A company expects the annual sales of one of their products next year to be 27 400 and anticipate
that the only significant changes in sales from one quarter to another to be those due to
anticipated seasonal effects.
a If there is no quarterly seasonal effect on sales of this product roughly how many of the
product should the company expect to sell each quarter?
b In fact the sales are seasonal in nature with the seasonal indices for each of the first three
quarters being as follows:
1st Quarter: 0.72 2nd Quarter: 0.88 3rd Quarter: 1.06
Predict the sales numbers for this product for each quarter of next year.
12 Calculate the total length of the minimum spanning tree for the network shown below and show
the minimum spanning tree on a diagram. (The numbers on each edge gives the length of that
edge in units.)
D
25 H
17 J
15
17
18 21
B 16 23
F
11 12
I
12
13 17
10 9
A
7
8 G
5
C E
13 A company identifies a number of tasks involved in the completion of a particular project. The
project network below shows these tasks labelled A to J with the numbers on each arc indicating
the number of days each task requires.
B5
D3
A7 C4 H4 I3 J2
Start Finish
E2
F5 G7
a Determine the minimum time to complete the project and the corresponding critical path.
b How many extra days could task E be allowed to take, over and above the 2 days already
allowed for, without delaying the completion time stated in part a?

14 The table below shows the number of gas units used by a company every two months for a period
of three years.
n Year Two-month period Units used

1 1 1st two months of year 1238
2 2nd two months of year 1812
3 3rd two months of year 5084
4 4th two months of year 6320
a Determine the seasonal indices for each two-month period of the year using the average
(mean) percentage method. Give answers in decimal form and correct to four decimal places.
b Use the seasonal indices to deseasonalise the number of units used for each two-month
period. Give each deseasonalised figure to the nearest ten units.
c View the deseasonalised figures plotted against n to check that linear regression would be an
appropriate model to use for this data pair.
d Use n and the deseasonalised data, D, to determine the least squares regression line D = an + b.
(Give a correct to two decimal places and b to the nearest ten.)
e Use your regression line, and the seasonal indices to reseasonalise the data, to predict the
number of units used for i the 1st two months of year 4,
ii the 4th two months of year 4.
f Suggest some events that could occur that would lead us to expect that the actual values for
year 4 could differ markedly from our predicted values.

15 A company is planning the construction of a new shopping complex.
The company identifies a number of tasks, the time each task will take and the order in which they
must be completed:
Task Time Order

A 1 week • Tasks B, C and D can commence provided task A has finished.
B 3 weeks • Task E can commence provided task B has finished.
C 1 week • Task F can commence provided task C has finished.
• Task H can commence provided task D has finished.
D 2 weeks
• Tasks K and I can commence provided task H has finished.
E 3 weeks
• Tasks G and M can commence provided both E and F have
F 4 weeks finished.
G 6 weeks • Task J can commence provided both G and I have finished.
H 5 weeks • Task L can commence provided both J and K have finished.
I 7 weeks
J 4 weeks
K 9 weeks
L 2 weeks
M 12 weeks
a Determine the minimum completion time and the critical path.

b What is the maximum time that task K could take without delaying the completion of
the project?
c If the time for task I could be reduced to 4 weeks what would be the minimum completion
time for the project?
16 A minimum spanning tree is needed for the network shown below. The number on each edge is
the cost, in $1000s, of making the connection. Find the minimum spanning tree for the network
and state its total cost.
F
E 30 35
I
40 50
H
B 38 45
45 30
20 34
C 12
A 30
G J
D 36 25
Before work is started on constructing the minimum spanning tree, it is found that a direct
connection could be made from H to I. What would the cost of this connection need to be for it
to be worthwhile to include in a new minimum spanning tree?

17 One particular day, three machines are to be assigned to manufacturing different products, with
each machine producing a different product. There are four products required but with just
the three machines working one product will not be manufactured that day. The assignment of
machine to product is to be done so that the total output, in total number of units produced for
the day, is maximised.
Clearly showing your use of the Hungarian algorithm, allocate machines to products to achieve
this, given that the numbers of units of each product that could be produced on each machine in
the day is as follows:
Product 1 Product 2 Product 3 Product 4

Machine A 2700 2300 3300 3300
Machine B 4200 3600 4100 4100
Machine C 4000 3500 3600 3300
18 A port authority wishes to build a railway network linking seven locations in the dockside complex.
The network is to be constructed so that it will be possible to travel by rail from any one of the
seven locations to any of the other six, perhaps not directly, but at least by going via other locations
in the network. The locations are Goods In, Goods Out, Waiting, Customs, Dock 1, Dock 2 and
Dock 3. The distances involved for the feasible rail connections from any of these locations to any
of the others are given in metres in the table below.
Goods In Goods Out Waiting Customs Dock 1 Dock 2 Dock 3

Goods In – – 420 – 300 – –
Goods Out – – – 150 – – 210
Waiting 420 – – 120 240 – –
Customs – 150 120 – – 210 180
Dock 1 300 – 240 – – 195 –
Dock 2 – – – 210 195 – 225
Dock 3 – 210 – 180 – 225 –
a Based on these distances determine the minimum length of track required for this network.
b In this ‘minimum length network’, which six pairs of locations would have a direct rail link?
(Direct rail link means able to journey from one location to the other without having to pass
through any of the other locations on the way.)
yawytra/moc.kcotsrettuhS

19 Five transport companies are asked to quote their price for four jobs involving the transportation
of wide loads. The company asking for the quotes is likely to have several more such jobs in
the future and wants to use a different company for each of the four jobs in order to see which
company seems to give the best service. The five transport companies are DG Transport, Haulage
and Co, JA Movers, Move it and Shift it and Relocate Plus. The prices each company quotes for the
four jobs are as in the following table:

DG Transport $18 000 $14 600 $28 000 $16 500
Haulage and Co $19 800 $12 800 $32 000 $18 000
JA Movers $18 000 $13 500 $27 000 $17 500
Move it and Shift it $24 000 $12 000 $25 000 $16 000
Relocate Plus $22 000 $13 000 $26 500 $17 000
Allocate a transport company to each job with a different company for each job, one company
missing out, and minimising the total cost of the four jobs according to the quotes given above.
Also state what this minimum cost will be.
yrarbilotohP /segamI ytteG

ANSWERS
Exercise 1A PAGE 10
1 a No, the increase cannot be attributed to the natural increase in the population. Numbers are given per 100 persons of
working age, thus the increase cannot simply be attributed to increase in population.
b Number of persons over age 65
40 per 100 persons of working age
30
20
10
Year
1971 1981 1991 2001 2011 2021 2031
c The number of persons over age 65 per 100 persons of working age shows an increasing trend with time. The
increase is steady at approximately 2 more people per 100, for each ten years from 1971 to 2011, and then increases
to about 5 more people per 100, for each ten years from 2011 to 2031.
2 a Line graph showing the population
of Western Australia, 1915 to 2005
2 200 000
2 000 000
1 800 000
1 600 000
noitalupoP
1 400 000
1 200 000
1 000 000
800 000
600 000
400 000
200 000 Year
5191
5291
5391
5491
5591
5691
5791
5891
5991
5002
b The population shows an increasing trend over the years from a population of just over 300 000 in 1915 to just over
2 million in 2005.
Earlier 10 year intervals in this time show an increase in population of about 60 000 people per ten years. Later 10
year intervals show an increase in population of about 300 000 people per ten years.
c The figures suggest that the population of Western Australia in 1950 was approximately 580 000. This prediction
involves interpolating between two known points (1945, 490 000) and (1955, 669 000). Hence the prediction should
be very reliable.
ISBN 9780170395069 Answers 173

d Continuing the trend suggests a population in 2025 of about
3 million (see graph on the right which shows two possible
continuations to the trend). However extrapolation is 3 000 000
involved as we are going beyond the known data points to
make our prediction. Continuing the line as shown is an
attempt to predict what will happen, but a lot could change 2 200 000
between the 2005 figure and 2025 so with this degree of 2 000 000
extrapolation involved the prediction should not be viewed 1 800 000
as being particularly reliable. If we had to predict it would 1 600 000
noitalupoP
perhaps be safer to predict in a range, say 2.8 million to 1 400 000
3.2 million. 1 200 000
1 000 000
800 000
600 000
400 000
200 000 Year
5191
5291
5391
5491
5591
5691
5791
5891
5991
5002
5202
3 a The number of short term visitors shows an increasing trend with time.
The increase seems reasonably steady from 1992–1997, Graph showing the number of short term
then the rate of increase slows, with numbers actually visitors arriving in Australia each year
declining in 2003 before picking up again. From 2005
to 2010 numbers of visitors are reasonably steady before 7 000000
increasing again from 2011 to 2013.
srotisiv mret trohs fo rebmuN 6 000000
5 000000
4 000000
3 000000
2 000000
1 000000
Year
2991
4991
6991
8991
0002
2002
4002
6002
8002
0102
2102
4102
b The graph suggest visitor numbers of approximately 4 600 000 for 1999.
This prediction involves interpolation so it should be reliable.
The flattening of the graph around 2005 to 2010 makes extrapolation to 2020 difficult to do with any confidence.
The 2013 figure perhaps suggests the upward trend continues so a prediction of approximately 7 500 000 for 2020
might be reasonable.
We are not extrapolating that far beyond the known values but extrapolation is involved so our prediction could not
be relied upon with any great confidence.

4 Graph showing the median age at first marriage
Age
30
29
28
27
26
25
24
23 Men
22 Women
21
20
Year
6691
8691
0791
2791
4791
6791
8791
0891
2891
4891
6891
8891
0991
2991
4991
6991
8991
0002
2002
4002
6002
8002
0102
2102
Written response not included here.
Compare your response with those of others in your class and discuss with your teacher.
5 N = 494 Y - 917 000
6 a Number of properties managed = 290 + 11.4t
b 655
7 a P = 0.8x + 17.2
b Percentage (P )
30
20
10
Year ( x)
5 10 15
Over the sixteen years the percentages of births delivered by caesarean section show
an increasing trend with most of the increase being from year 5 to year 11.
For the first five years the percentage remains level at around 18%.
For the last four years the percentage remains level at around 30%.
8 a Average age of vehicles
12
11
sraey ni egA
10
7
Year
1990 2000 2010 2020
ISBN 9780170395069 Answers 175

b The data is not suitable for linear regression. The graph Average age of vehicles
shows that the distribution of points does not follow 12
a linear pattern, nor anything that could be well
approximated by a linear pattern. 11
sraey ni egA
Predicted values given by the ‘best straight line’ as 10
determined using linear regression would overestimate
ages for the early and late years, and underestimate 9
ages for the middle years – see diagram on the right
in which the least squares linear regression line is 8
shown as a broken line.
7
Year
1990 2000 2010 2020
c Predicted average age of vehicle for 1999 would be Average age of vehicles
about 9.9 years. We have data for 1998 and 2000 12
so our value for 1999 involves interpolation.
It should therefore be a prediction that could 11
be relied on with confidence.
sraey ni egA
10
We cannot predict with any great confidence what
the average age of vehicles will be in 2020 as 2020 9
is well beyond the years plotted (i.e. extrapolation
is involved). If decline in later years were to continue, 8
the figure for 2020 might perhaps be around 7.8.
No great confidence in reliability of this 7
prediction though. Year
1990 2000 2010 2020
9 Viewing the graph of Y plotted against t confirms the unsuitability of linear regression.
Y = 4.69 × t 2.145 .
Miscellaneous exercise one PAGE 16
1 $2316.50
2 $3276
3 A: y=x B: y = 2x + 1 C: y = 0.5x + 1 D: x = 3 E: y = 2 F: y 2x - 3
= -
4 a T1 = 25, T2 = 27, T3 = 29, T 4 = 31, T5 = 33

b T1 = 32, T2 = 30.5, T3 = 29, T 4 = 27.5, T5 = 26
c T1 = 64, T2 = 96, T3 = 144, T 4 = 216, T5 = 324
d T1 = 4000, T2 = 2000, T3 = 1000, T 4 = 500, T5 = 250
e T1 = 5, T2 = 17, T3 = 41, T 4 = 89, T5 = 185
f T1 = 16, T2 = 20, T3 = 26, T 4 = 35, T5 = 48.5

5 Percentage of WA population living in Perth
Year 1911 1921 1933 1947 1954 1961 1966 1971 1976 1981 1986 1991
% 41% 51% 52% 60% 62% 65% 67% 68% 71% 71% 72% 73%
100% Population of Perth as a percentage Perth population, as % of WA population, shows

90% of WA population an increasing trend.
80%
This Increase ‘levels off’ in later years but the
70%
60%
percentage is still increasing.
50% b i For 1940 approx. 56%.
40% 1940 is between years for which the population
30%
is known so interpolation is involved.
20%
10% Estimate should be reliable.
Year
ii For 2020 approx. 80%.
0191
0291
0391
0491
0591
0691
0791
0891
0991
Extrapolation is involved and 2020 is a long way
beyond known values. Estimate may not be too
reliable.
(Note: Linear regression gives 86% but not
a particularly suitable model.)
Exercise 2A PAGE 23
1 3-point 2 4-point 3 4-point 4 3-point 5 5-point 6 3-point

7 a and b See graph: Attendance
500
400
300
200
100
Performance number
5 10 15
c Many factors such as cost, venue and actor availability, would need to be considered before we could decide if a sixth
week would be ‘a good idea’. However, attendance figures suggest a 6th week could possibly have given attendance
similar to week 1 so on that basis alone, and if the attendance of week 1 was considered acceptable, a 6th week could
have been a good idea.
ISBN 9780170395069 Answers 177

8 a Four-point moving averages are:
26.5, 26.25, 24.5, 24.5, 24, 23.5, 23.5, 23, 22.25, 22, 20.75.
b Number of pigs
35
30
25
20
15
10
5
Year 1 Year 2 Year 3
1st 2nd 3rd 4th 1st 2nd 3rd 4th 1st 2nd 3rd 4th 1st 2nd
Q Q Q Q Q Q Q Q Q Q Q Q Q Q
c Decreasing.
9 a A = 263, B = 270, C = 276, D = 277, E = 256.

b Bill ($)
350
300
250
200

150
100
50
2012 2013 2014

1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
Two month interval
c Increasing.
Exercise 2B PAGE 29
1 Consumption of soft drink could well be seasonal in nature. Hence, with the data involving consumption per quarter
year a 4-point average would be most suitable.
a A = 6940, B = 6892, C = 6788, D = 6763, E = 5511, F = 6638, G = 6662
2 a A = 176.5, B = 180, C = 179, D = 182, E = 183, F = 184, G = 181,
H = 190, I = 189, J = 191.5

3 Year 1 Year 2
Summer Autumn Winter Spring Summer Autumn Winter Spring
1228 364 640 1220 1436 276 752 1132
4-point MAs 863 915 893 921 899
4-point CMAs 889 904 907 910
4 a We would expect the production of fruit to be dependent on the weather and hence follow a seasonal pattern. Hence
determining a seasonal moving average makes sense and with data collected on a monthly basis a 12-point moving
average is sensible.
b Price per kg
$5.00
$4.50
$4.00
$3.50
$3.00
$2.50
$2.00
$1.50
12-point centred moving average
$1.00
$0.50
Year 1 Year 2 Year 3 Year 4

J FMAM J J A S OND J FMAM J J A S OND J FMAM J J A S OND J FMAM J J A S OND
c Underlying trend shows a slight increase over time.

5 a The 5-point moving average would smooth the data more than the 3-point moving average.
b 1500 Number of incidents c In the years under consideration
1400 there is an initial downward trend
1300 followed by an upward trend, but
1200 then from about the year 2000 the
1100
trend has been decreasing.
1000
900 Hence the expectation mentioned
800 at the beginning of the question,
700 i.e. that with increasing population
600 we would expect the number of
500 criminal incidents to increase, was not
400 the case for this category of crime.
300 5-point moving average
200 (From 1990 to 2013 the NSW
100 population increased from
Year
5.8 million to 7.4 million.)
0991
2991
4991
6991
8991
0002
2002
4002
6002
8002
0102
2102
ISBN 9780170395069 Answers 179

Exercise 2C PAGE 36
1 The seasonal index for the fourth quarter is 98%.

2 The seasonal index for Wednesday is 94%.
3 a With no seasonal effect on sales the company should expect to sell roughly 13 000 each month.
b Predicted sales for each month would be as follows:
January 11 440
February 10 660
March 9880
April 9 620
May 10 790
June 12 740
July 13 520
August 13 650
September 14 040
October 14 430
November 16 380
December 18 850
4 The figures suggest that sales for the whole year will be 62 000.
For each of the remaining quarters the figures suggest sales will be:
2nd quarter 16 120, 3rd quarter 15 190, 4th quarter 13 330.
5 Week Period of time Attendance Daily mean Attendance as % of

for the week week’s daily mean
1 Friday 2346 2766 84.82%
Saturday 3143 113.63%
Sunday 2809 101.55%
2 Friday 2572 2929 87.81%

Saturday 3258 111.23%
Sunday 2957 100.96%
3 Friday 2987 3231 92.45%

Saturday 3500 108.33%
Sunday 3206 99.23%
Friday Saturday Sunday

1st week 84.82% 113.63% 101.55%
2nd week 87.81% 111.23% 100.96%
3rd week 92.45% 108.33% 99.23%
Index (1 dp) 88.4% 111.1% 100.6%

6 Year Period of time Units sold Quarterly mean Units as % of year’s
for the year quarterly mean
1 1st quarter 54 000 52 650 102.56%
2nd quarter 63 200 120.04%
3rd quarter 45 200 85.85%
4th quarter 48 200 91.55%
2 1st quarter 48 500 47 250 102.65%

2nd quarter 56 700 120.00%
3rd quarter 41 300 87.41%
4th quarter 42 500 89.95%
3 1st quarter 51 800 49 675 104.28%

2nd quarter 62 600 126.02%
3rd quarter 40 500 81.53%
4th quarter 43 800 88.17%
1st quarter 2nd quarter 3rd quarter 4th quarter

1st year 102.56% 120.04% 85.85% 91.55%
2nd year 102.65% 120.00% 87.41% 89.95%
3rd year 104.28% 126.02% 81.53% 88.17%
Seasonal index (1 dp) 103.2% 122.0% 84.9% 89.9%
7 n Week Day No. of calls 5-pt MA Calls as % of daily

average for week
1 Mon 258 – 107.59%
2 Tue 231 – 96.33%
3 One Wed 215 239.8 89.66%
4 Thur 248 235.6 103.42%
5 Fri 247 231.0 103.00%
6 Mon 237 224.8 109.32%
7 Tue 208 221.6 95.94%
8 Two Wed 184 216.8 84.87%
9 Thur 232 215.8 107.01%
10 Fri 223 213.0 102.86%
11 Mon 232 212.8 113.06%
12 Tue 194 208.2 94.54%
13 Three Wed 183 205.2 89.18%
14 Thur 209 – 101.85%
15 Fri 208 – 101.36%
ISBN 9780170395069 Answers 181

Week 1 Week 2 Week 3 Seasonal index
Monday 107.59% 109.32% 113.06% 109.99%
Tuesday 96.33% 95.94% 94.54% 95.60%
Wednesday 89.66% 84.87% 89.18% 87.90%
Thursday 103.42% 107.01% 101.85% 104.09%
Friday 103.00% 102.86% 101.36% 102.41%
Exercise 2D PAGE 43
(You may occasionally find that an answer you obtain varies slightly from the answer given here depending on when, and to
what degree, rounding occurred.)
1 $22 300 ( = $25 400 ÷ 1.14 and then given to the nearest $100).
2 173 700 units ( = 132 000 ÷ 0.76 and then given to nearest 100).
3 a 69 530 people. b 69 940 people.
4 The analysis predicts that 14 750 units will be sold in January.
5 The figures suggest that the actual weight of the fruit this summer will be 23 680 kg.
6 a Seasonal index for Jan- Apr: 107.25%
Seasonal index for May- Aug: 75.56%
Seasonal index for Sept- Dec: 117.20%
b Jan–Apr May–Aug Sept–Dec
Year 1 1.71 1.71 1.69
Year 2 1.74 1.87 1.76
Year 3 1.80 1.69 1.84
Year 4 2.03 2.01 1.99
7 a Seasonal index for Friday: 0.6436 Seasonal index for Saturday: 1.0540
Seasonal index for Sunday: 1.1555 Seasonal index for Monday: 1.1469
Friday Saturday Sunday Monday
Year 1 14 290 14 080 14 070 14 690
Year 2 15 240 14 980 14 950 15 520
Year 3 14 680 14 440 14 890 14 660
Year 4 15 290 16 030 15 610 14 600
8 a Monday Tuesday Wednesday Thursday Friday Saturday Sunday

Week 1 $652 $643 $690 $688 $644 $645 $721
Week 2 $734 $760 $760 $787 $746 $707 $725
Week 3 $855 $837 $779 $751 $852 $895 $778
b Graph of deseasonalised data, D, plotted D
against t, as shown on the right, shows $1400
a reasonable linear trend. Hence linear
regression is appropriate. $1200
$1000
$800
$600
$400
$200
5 10 15 20 t

c D = 9.64t + 639.1
d For Monday of week 4 (t = 22) the regression line predicts value for D of $851.2.
Hence prediction of real takings will be $851.2 × 0.8242 = $702, to the nearest dollar.
For Wednesday of week 4 (t = 24) the regression line predicts value for D of $870.5.
For Saturday of week 4 (t = 27) the regression line predicts value for D of $899.4.
9 a Viewing the graph, not shown here, confirms reasonableness of using linear regression for (n, M) data.
b M = 1.0994n + 26.031
c Monday Tuesday Wednesday Thursday Friday
Week one 1.151 32 0.822 37 0.789 47 0.921 05 1.315 79
Week two 1.075 58 0.872 09 0.784 88 0.872 09 1.395 35
Week three 1.144 28 0.796 02 0.820 90 0.870 65 1.368 16
Week four 1.043 48 0.913 04 0.869 57 0.956 52 1.217 39
Mean (4 dp) 1.103 7 0.850 9 0.816 2 0.905 1 1.324 2
d Missing figures are: Tuesday Wednesday Thursday

Week one 29 29 31
Week two 35 33 33
Week three 38 40 39
Week four 49 49 49
e Predictions for week 5: Mon 54 Tues 43 Wed 42 Thur 47 Fri 71
f If week five involved things like ‘pupil free days’, public holidays, unusually severe weather making travel to school
difficult, a flu epidemic in the area, school closure due to fire etc., our week five attendance predictions would not be
reliable. (Also, we cannot assume that the trends of the first four weeks will necessarily continue even without these
unusual events.)
10 a Rescues as a percentage of annual mean

(As percentages and correct to 2 decimal places)
Oct/Nov Dec/Jan Feb/Mar
Season 1 56.10% 135.37% 108.54%
Season 2 47.37% 136.84% 115.79%
Season 3 51.43% 152.86% 95.71%
Season 4 53.13% 142.19% 104.69%
Seasonal indices
52% 142% 106%
(Nearest percent)
b Oct/Nov Dec/Jan Feb/Mar

Season 1 88 78 84
Season 2 69 73 83
Season 3 69 75 63
Season 4 65 64 63
ISBN 9780170395069 Answers 183

c Viewing the graph, not shown here, confirms reasonableness of using linear regression for (n, D) data.
Using integer values of D: D = 85.83 - 2n.
(Using more accurate values of D: D = 86.05 - 2.003n.)
d n Season Months Predicted number of rescues
13 Oct/Nov 31
14 5 Dec/Jan 82
15 Feb/Mar 59
Miscellaneous exercise two PAGE 48
1 $247.50
2 1
3 A seasonal index of 0.87 for February means that whatever it is that we are measuring for February tends to be
13% below the monthly average.
4 Season Spring Summer Autumn Winter

Deseasonalised data 1000 1552 1600 1875
5 The starting amount is $1 500 000.

6 The scattergraph indicates that in general the more p
recently manufactured the vehicle the higher the 60 000
asking price. Given the number of points involved,
and their distribution, the relationship between 50 000
)$( ecirp gniksA
year of manufacture and asking price could be

40 000
well represented by a linear model.
p = 3561t - 697 30 000
Likely price for a 2007 model: $31 350.
20 000
10 000
t (t = 0 is year 1998)
5 10 15
7 a The given revenues have a total of $3 760 000 and thus a quarterly average of $940 000. The first quarter is
20% down on this quarterly average (compared to the 15% down the seasonal index suggests), the second quarter
is 21% up (compared to the expected 25%), the third quarter is 25% up (compared to the expected 12%) and the
fourth quarter is 26% down (compared to the expected 22%). Thus the third quarter revenue is a surprise when
compared to what we might expect from the seasonal indices.
b If all other factors remained the same then the third quarter increase in the revenue share above expectation could
be as a result of the advertising campaign being successful.

8 a t Year Quarter Number sold Number sold as Number sold
in quarter percentage of quarterly seasonally adjusted
mean for the year
1 One 1st 1637 106.99% 1514
2 2nd 1489 97.32% 1537
3 3rd 1244 81.31% 1598
4 4th 1750 114.38% 1494
5 Two 1st 1405 108.16% 1299

6 2nd 1241 95.54% 1281
7 3rd 1012 77.91% 1300
8 4th 1538 118.40% 1313
9 Three 1st 1253 109.24% 1159

10 2nd 1121 97.73% 1157
11 3rd 852 74.28% 1095
12 4th 1362 118.74% 1162
Year 1 Year 2 Year 3 Seasonal index

1st Quarter 106.99% 108.16% 109.24% 108.13%
2nd Quarter 97.32% 95.54% 97.73% 96.86%
3rd Quarter 81.31% 77.91% 74.28% 77.83%
4th Quarter 114.38% 118.40% 118.74% 117.17%
c Using integer values for seasonally adjusted figures: S = -43.885t + 1611

(Using more accurate values for seasonally adjusted figures: S = -43.887t + 1611.1)
d Predicted numbers sold for year 4 are:
1st quarter 1125, 2nd quarter 965, 3rd quarter 742, 4th quarter 1065.
Exercise 3A PAGE 56
1 $2720 2 5 years 3 12.5% 4 $2400 5 7.4% 6 $7700

7 $6347.50 to the nearest cent. 8 $9609.32 to the nearest cent.
9 Annual compounding gives interest of $13 123.85, to the nearest cent.
Monthly compounding would give $400.61 more interest, to the nearest cent.
10 Annual compounding gives interest of $463 570.00, to the nearest cent.
Daily compounding would give $26 568.34 more interest, to the nearest cent.
11 $6 965.59
12 It would take approximately 11.64 years for the investment to double in value. I.e. approximately 11 years and 8 months.
13 It would take almost 20 compoundings (19.89), i.e. just under 10 years (9.945 years).
ISBN 9780170395069 Answers 185

1st term 2nd term 3rd term 4th term 5th term 15th term
14 5, 8, 11, 14, 17. 47.
15 5, 2, - 1, - 4, -7. -37.
16 - 10, - 5, 0, 5, 10. 60.
17 12.5, 15, 17.5, 20, 22.5. 47.5.
18 0.25, 0.5, 1, 2, 4. 4096.
19 20 480, 30 720, 46 080, 69 120, 103 680. 5 978 711.25.
20 2 621 440, 1 310 720, 655 360, 327 680, 163 840. 160.
30 29 28 27 26
21 2 , 2 , 2 , 2 , 2 . 2 16.
22 5, 13, 29, 61, 125. 131 069.
23 $1000, $1300, $1660, $2092, $2610.40. $18 758.78.
24 $5000, $5000, $5000, $5000, $5000. $5000.
25 12, 23, 34, 45, 56. 166.
26 819 200, 819 200, 819 200, 819 200, 819 200. 819 200.
27 1, 1, 2, 3, 5. 610.
28 2, 3, 8, 19, 46. 309 268.
29 5, - 5, - 5, 5, 5. - 5.
30 5, - 3, - 1, 3, 5. - 1.
Exercise 3B PAGE 59
Percentages given rounded to three decimal places where rounding is necessary.
1 Compounding Nominal annual Effective annual

Annual 4% 4%
Six monthly 4% 4.04%
Quarterly 4% 4.060%
Monthly 4% 4.074%
Weekly 4% 4.079%
Daily 4% 4.081%

Annual 8% 8%
Quarterly 8% 8.243%
Monthly 8% 8.300%
Weekly 8% 8.322%
Daily 8% 8.328%
3 The effective annual interest rates for 8% are not simply double the rates for 4%.
in
Doubling the value of ‘i’ in the formula  1 +

- 1 will not simply double the answer.
 n 

Exercise 3C PAGE 63
1 Tn + 1 = 1.08 × Tn + $200 $11 533.01 is in the account at the end of ten years.
2 a Tn + 1 = 1.005 × Tn + $100 b Final value of account is $9817.01. (= T36 - $100)
3 $22 474.58 (nearest cent). 4 $4780.04 (nearest cent). 5 $14 773.77 (nearest cent).
6 a 10.164% (correct to three decimal places). b 7.590% (correct to three decimal places).
Exercise 3D PAGE 65
1 a End of first year $195 000 b End of first year $187 000
End of second year $170 000 End of second year $158 950
End of third year $145 000 End of third year $135 107.50
2 a First year depreciation is $10 000 b First year depreciation is $16 000
Second year depreciation is $10 000 Second year depreciation is $12 800
Third year depreciation is $10 000 Third year depreciation is $10 240
3 a $23 000 b $16 000 c $9000
Exercise 3E PAGE 67
1 A: $182.738 9691 B: $182.74 C: 12 D: $4 000 E: $0

F: 2 G: 9 H: monthly I: 12
2 A = 5359, B = 3 894.90, C = 512.83, D = 564.11.
3 A = 1.015, B = 1043.86, C = 180.64.
4 $49 700, $49 397, $49 090.97, $48 781.88 (nearest cent).
5 a At the end of year five and immediately after his regular monthly repayment, Hisham still owes $15 016.34 on the loan.
b Hisham would owe $12 974.59 at the end of the 5 years (after his repayment for the month).
6 a Pays off the loan at the end of the 70th month (and the 70th repayment is less than $250).
b Pays off the loan at the end of the 56th month (and the 56th repayment is less than $290).
c Pays off the loan at the end of the 48th month (and the 48th repayment is less than $290).
7 The constant monthly repayment needs to be $443.21, to the nearest cent.
8 The constant monthly repayment needs to be $717.35, to the nearest cent.
9 The constant monthly repayment needs to be $1 972.11, to the nearest cent.
10 The loan would be paid off in exactly two years by paying $188.29 per month for 24 months (actually a few cents more
on the very last repayment to exactly pay off the outstanding amount).
If instead the monthly interest rate was 1.2% per month the monthly repayments would need to be $192.81 (now a few
cents less on the very last repayment).
ISBN 9780170395069 Answers 187

Exercise 3F PAGE 72
1 Loan of $514 214.53 + deposit of $17 000.

Hence, to the nearest $1000, Fran and Michael can afford a house costing $531 000.
Hence, to the nearest $1000, Peta and Peter can afford a house costing $326 000.
Hence, to the nearest $1000, Rania and Umah can afford a house costing $476 000.
Hence, to the nearest $1000, Hue and James can afford a house costing $684 000.
Hence, to the nearest $1000, Kirra can afford a house costing $395 000.
6 $3 575.13 (Note: Round down from $3 575.130 92. Final payment will be a little over $3575.13.)
$622 540 (nearest $10) Note: 300 × $3575.13 - $450 000 gives $622 539.
Allowing for larger 300th payment gives $622 539.93.
Different rounding regime, e.g. on some online amortization calculators, gives $622 539.38.
7 $2 991.93 (Note: Round up from $2991.927 336. Final payment will be a little under $2991.93.)
$718 060 (nearest $10) Note: 240 × $2991.93 gives $718 063.20.
Allowing for reduced 240th payment gives $718 061.78.
Paying $100 more per month would pay off the loan approximately 16 and a half months earlier.
Miscellaneous exercise three PAGE 72
1 a $11 754.62 b $11 887.58 c $11 918.77

2 A seasonal index of 1.08 for Autumn means that whatever it is that the index is referring to, e.g. sales numbers, people
moving interstate, unemployment figures, etc, the measurements for Autumn tend to be 8% above the average for
a season.
3 Season Mon Tue Wed Thur Fri Sat

Deseasonalised data 123 113 113 125 117 133
4 a T1 = 7, T2 = 19, T 3 = 31, T4 = 43, T5 = 55

b T1 = 100, T2 = 85, T 3 = 70, T4 = 55, T5 = 40
c T1 = 5000, T2 = 6000, T 3 = 7200, T4 = 8640, T5 = 10 368
d T1 = 2000, T2 = 200, T 3 = 20, T4 = 2, T5 = 0.2
e T1 = 4, T2 = 11, T 3 = 25, T4 = 53, T5 = 109
f T1 = 200, T2 = 296, T 3 = 440, T4 = 656, T5 = 980
5 a D b B, C c A d B, D e B
6 a $48 620.70 b $45 810.01 c $41 460.41 d $32 257.88
7 a $16 483.73
b 15.19 years, i.e. just over 15 years and 2 months.
c 18 years.
d i at least 11.3264% ii at least 10.7777%

8 Number of yen purchased for 1 A$
450
400
350
300
250
Raw data
200
150
100
50
Year
0
1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Comments summarising the data, and about today’s exchange rate, are not included here. Compare your comments to
those of others in your class.
Exercise 4A PAGE 81
1 If $8000 (or more) is invested at 7.5% per annum then the interest earned will be $8000 × 0.075 i.e. $600 (or more).
Hence the interest each year will equal (or exceed) the award amount. The balance in the account will then remain at
(or exceed) $8000 thus allowing the award to continue ‘in perpetuity’.
2 1.024 = 1.082 432 16. Thus 8% per annum, with quarterly compounding, increases an initial investment by 8.243 216%.
If the initial investment is $A we require 8.243 216% of $A to equal $75 000.
Thus $A = $909 839.0725, i.e. $909 840 rounded up to the next dollar.
The one off investment needs to be $909 840 (or more).
(Other approaches are possible, for example trial and adjustment.)
3 $243 199.29 (or more).
4 a Increase P. b An increase in R. c A decrease in A .
5 There will be $436 670 left in the account after the tenth withdrawal (nearest ten dollars).
Kelvin can withdraw $50 000 per year for 22 years. (Then, at end of the 23rd year, the withdrawal would be less than
$50 000 and would close the account.)
Had the interest rate been 7.8% Kelvin would have been able to withdraw $50 000 per year for 45 years. (Then, at the
end of the 46th year, the withdrawal would be less than $50 000 and would close the account.)
ISBN 9780170395069 Answers 189

6 Julie will be able to withdraw $50 000 from the account for 26 years. (Then, at the end of the 27th year, the withdrawal
would be less than $50 000 and would close the account.)
If instead Julie withdrew $45 000 per year she could do this for 34 years. (Then, at the end of the 35th year, the
withdrawal would be less than $45 000 and would close the account.)
To maintain her standard of living it is likely that Julie will need an increasing amount to live on each year because of
the likely rise in the cost of everything due to inflation. Hence she might be wiser to withdraw an amount which
increases each year to allow for this inflation, even if it means having to withdraw less than the $50 000 in the early
years.
7 $352 941 8 $575 251 9 $358 663 10 $454 576
11 a and b Nineteen annual payments of $30 000 and then at the end of 20 years the final payment will be $5214.89
(= $4851.06 × 1.075).
12 a and b One hundred and seventy three monthly payments of $4000 and then at the end of 174th month the final
payment will be $2534.41 (= $2524.31 × 1.004).
13 a and b $381 236
14 a and b $46 732
15 a Immediately after the third withdrawal the balance in the account will be $277 722.60.
b Immediately after the 14th withdrawal the balance is $26 395.75. Hence at the end of the 15th year, by withdrawing
$28 507.41 (= $26 395.75 × 1.08) the balance will be reduced to zero.
16 a Immediately after the eighth withdrawal the balance in the account will be $149 506.05.
b Immediately after the 13th withdrawal the balance is $575.01 ($575.009). Hence could simply add this to the
payment at the end of the 13th year and reduce the balance to zero or could continue for another year and at the
end of the 14th year, by withdrawing $618.13 (= $575.009 × 1.075) the balance will be reduced to zero then.
Exercise 4B PAGE 87
1 $259 162.18 2 $567 873 3 $333 114.12 4 $565 537.24

5 Fifty seven quarterly payments of $15 000 followed by a final payment of $9924.23 which closes the account.
Computer spreadsheet or graphic calculator activity PAGE 88
Answers not given here. Compare your answers with those of others in your class.
Miscellaneous exercise four PAGE 89
1 $2448.90
2 The 855 is the gradient of the line of best fit. It informs us that over the ten years under consideration the number of
people attending the music festival increased at approximately 855 people per year.
3 a 6.136% b 6.168% c 6.183%
4 $4267.45
5 a $4300 b 6.5% c $6274.31 d 20.795%
6 The initial investment was $1600, at a fixed monthly interest rate of 1.5% with a fixed monthly deposit of $250.

7 a A: 0.7570, B: 0.8033, C: 0.7778, D: 0.7991, E: 160, F: 0.7625, G: 1.175
b Calculation of seasonal indices
2010 2011 2012 2013 2014 Seasonal
index (3 dp)
1st 4 Months 1.1190 1.0233 1.1264 0.9953 1.0625 1.065
2nd 4 Months 0.7570 0.8033 0.7778 0.7991 0.7625 0.780
3rd 4 Months 1.1241 1.1733 1.0958 1.2056 1.1750 1.155
Seasonally adjusted sales figures, S (Nearest integer)

2010 2011 2012 2013 2014
First 4 Months 415 288 276 200 160
Second 4 Months 383 309 260 219 156
Third 4 Months 384 305 248 223 163
c Using integer values for S: S = -17.91t + 409.2

(Using more accurate values for S: S = -17.919t + 409.36)
d For the first 4 months of 2015 predicted sales are 131 (nearest integer).
For the second 4 months of 2015 predicted sales are 82 (nearest integer).
For the third 4 months of 2015 predicted sales are 100 (nearest integer).
Exercise 5A PAGE 96
188 units
2
1920 units
137 units
122 units
ISBN 9780170395069 Answers 191

5
or
207 units 207 units

6
or
2250 units 2250 units

7 Replace AG, GF, FC, CB, FE, ED with new piping. Close off AF, AC, AB, CE, CD, BD. New piping has a total length
of 325 m.
8 295 m A2
B1
P2 P1
35 30
40 40
60 50
A1 B2 A3
40
9 a C D
B
E
G
A
F
b The total length of tunnelling that will be closed off to visitors is 445 metres.
10 a Graph showing roads and distances not given here. b 1890 km
Exercise 5B PAGE 102
1 76 units 2 129 units 3 1680 units

4 19.7 units 5 2520 units 6 25.8 units, AB, BF, CF, DE, EF
7 21.1 km 8 $670
9 a 921 km
b Harlow - Raine, Wayley - Gatley, Wayley - Olber, Raine - Lewis, Raine - Fitch, Fitch - Wayley
10 BF, CH, DF, EA, FE, GD, HG, IC, JC. 261 km.

Miscellaneous exercise five PAGE 105
1 a N = 29.14n + 7668
b The value of a, 29.14, means that linear modelling of the data suggests that during the time period the data covers
the number of people employed full time is increasing at the rate of approximately 29 140 people per quarter.
c July 2011 has an n value of 7. The line of regression gives N = 7872.
Linear modelling gives a predicted employment figure for July 2011 of 7872 thousand.
January 2015 has an n value of 21. The line of regression gives N = 8280.
Linear modelling gives a predicted employment figure for January 2015 of 8280 thousand.
We would expect the July 2011 prediction, which involves interpolation, to be more reliable than the January 2015
prediction, which involves extrapolation. The latter involves predicting beyond known data points and events could
happen in the meantime making such predictions unreliable.
d Viewing the data graphically, as shown on the right, suggests N
that linear modelling may be unwise. The overall trend seems 8100
more curved than straight. We could perhaps use two linear
models but even then the apparent late downturn in employment 8000
suggests that extrapolating ahead of the given data based on
7900
a linear model could well be unwise.
The placement of the linear regression line shows why the 7800
predicted figure for July 2011 was below the real figures on
either side of it. Hence using a linear model for this data even 7700
made an interpolated prediction doubtful.
7600
7500
4 8 12 n
Jan Jan Jan Jan

2010 2011 2012 2013
2 At 6% per annum, i.e. 0.5% per month, the initial $350 000 will earn interest of $1750 by the end of the first month.
Hence a withdrawal of $1500 will leave more than $350 000 in the account for the next month. The account will
continue to grow each month and the task set for the calculator, to find how many payments it takes to give a final
balance of zero, is impossible. Hence the ‘Error’ response.
3 a = 67, b = 76, c = 80, d = 71.5, e = 81, f = 76, g = 80, h = 78,
i = 69, j = 82, k = 80, l = 86, m = 84, n = 82, o = 53.
4 $3634.60
5 a At the end of year six and immediately after her regular repayment, Amy still owes $5636.26 on the loan.
b To pay off the loan in exactly six years Amy’s annual repayments need to be $3264.44.
Exercise 6A PAGE 113
1 60 units 2 90 units 3 80 units 4 100 units 5 450 units

6 550 units 7 500 units 8 170 units 9 550 units 10 100 units
11 Max flow is 850 units. Q U
200
One way it can be achieved is as shown.
350 200
150 T
200
150 50
P
V
250
350 150 S
200
ISBN 9780170395069 Answers 193

12 v = 100, w = 300, x = 100, y = 200, z = 150.
13 1800 B 1000 D
800 200 200 1200

600 C 200
A E
400
14 16 units
Upgrade AC to 8 units: Increase the maximum flow to 19.
Upgrading BD or FG: Will not alter the maximum flow.
1 I: 150 units, II: 130 units, III: 160 units, IV: 140 units, V: 180 units.
2 450 3 4
500
550
5 170 6 7
100
550
Miscellaneous exercise six PAGE 117
1 As percentages, and to the nearest percent, the seasonal indices for the first, second, third and fourth quarters are 141%,
112%, 59% and 88% respectively.
Numbers for the first quarter tend to be 41% up on the quarterly average, for the second quarter 12% up on the
quarterly average, for the third quarter 41% down on the quarterly average and for the fourth quarter 12% down on the
quarterly average.
2 241 units
3 a $4 026.14 needs to be repaid per month for the loan to be paid off in 25 years.
(Note: Rounding up from $4026.135 417 so final payment will be a little under $4026.14.)
b A total of $707 840 (nearest $10) interest is paid during the 25 years.
Note: $4026.14 × 12 × 25 - $500 000 gives $707 842.
Allowing for reduced 300th payment gives $707 837.27.
4 a $12 000 b 7.2% c $18 211.68
5 $14 400 in first year, $12 672 in second year, $11 151.36 in third year.
6 1250 L/ minute. 200
B C
(The maximum flow is 1250 L/min but
the maximum flow diagram shown here is 400 300
not the only possible way of achieving it.) 600 200 700
150 50
D 250 E
A F
500

7 a Three-point moving averages are b
shown below. (Values are given to 2 500 000
the nearest 100, as requested.)
n 3-pt MA (M)
2 000 000
srotisiv mret-trohs fo rebmuN

1 –
2 1 837 500
3 1 828 600 1 500 000
4 1 794 000
5 1 830 100
6 1 850 500 1 000 000
7 1 894 900
8 1 930 000
500 000
9 1 938 800
10 1 937 900
n
11 1 923 600
3 6 9 12 15 18
12 1 944 700
13 1 964 700
14 2 010 700
15 2 037 400
16 2 076 800
17 2 126 800
18 –
c M = 18 970 n + 1 753 000

d Interpretation of the number 18 970 in the equation M = 18 970n + 1 753 000:
The number of short-term visitors arriving in Australia from January 2008 to December 2013 was, on average,
increasing at a rate of approximately 19 000 people per four months.
1 a 24 days b PTUWY c 3
2 a 12 hours b AEFH c 12 hours (i.e. no change) d 1
3 a 17 days b ACHLN c 17 days (i.e. no change)
4 a 27 hours b PSVZ c No (cut by 1 hour, P W X Y becomes critical)
d P
5 a 30 days b ACGJ c 14 days into the project. d 16 days into the project.
e 2 days f 4 days g 3 days h F
6 a 74 hours b PQRS c 10 hours d 29 hours
e 19 hours f 57 hours g 10 hours
7 a 17 minutes, P T Y and S X Z b 2 minutes each.
8 a 44 days b BCGJ c 14 days d 2 days
e 4 days
9 a 3 h 10 min b 12:10 p.m. c 11:05 a.m.
10 a 4:25 p.m. b 4:55 p.m. c 45 mins
ISBN 9780170395069 Answers 195

1 a R3 2 a
Q6 S5
P7 S4 W7 T4
Q5 V5 X2 X4
W2
U3 Z5
R8 Y3
T6 U9
V7
b 19 weeks, P - R - W - X b 21 days, R - U - W - Z - Y
c 4 weeks c 23 days
3 a 4 a E15
A16 B21
A10
B2
D25 G16 1
F11
C28 C16
H17
E5 G17
F25
D35
b 78 minutes, A - B - D - G b 52 days, C - E - B
c 3 minutes c 8
5 a E10 6
G5 Y4 R10
A6 F3 Q6
D7 B7 P5 T4 S2 V4
Z4
I4
W3
H9 C4 X10 U9
b 21 hours, A - E - G 27 minutes, P - W - X - U
c D and H
7 a W17 8 C45 F45 I45

R16 S22 X15 A30 B5 D30 G30 J30 L30 M2
Y13
U20 Z18
E30 H30 K30
T25 V28
b 79 days, T - S - W-X 11:50 a.m.

c 80 days
9 a W14
Q4 R4
X7
S7 U3
Y9 Z3
T9
V11
b 26 hours, T - U - W
c i Two hours over the allocated three.
ii Three hours over the allocated three.

10 a F2
A3 G3
D4
B3 C1 E1 H2 I1
b 16 days, B - C - D - F - G - H - I
c 8 days
d Reduce by one day.
11 F15
B15 G25
C8 H12
A5 D5 I5
E10
J8 K8
Minimum completion time 60 minutes.

Attempt to speed up:
Service vehicles getting in position, passengers on and off, security check.
12
A5 D6
Q0
B4 C7
Critical path is A - Q - C
Minimum completion time is 12 hours.
13 F1 G1
A10 H1
B3 C1 I3 J2
D3
E1
Minimum completion time is 18 weeks.
Miscellaneous exercise seven PAGE 136
1 A seasonal index for January of 1.12 means that whatever it is that was used to determine the index, e.g. sales numbers,
gross takings, number of people, prices etc., increases in January by 12% of the monthly average.
2 Five years after the commencement of the loan the amount owed is $243 862.70 (to the nearest cent).
3 Seasonally adjusted figures are 2208 for January and 2269 for February.
4 a A, C. b B, D. c A, C, D. d C e D
5 The maximum number of units that can be delivered is 750.
Increasing the capacity of AC to 250 units increases the max flow by 50 units (to 800 units).
6 Connections need to be: A to B, B to Base, Base to D, D to C. $460 000
7 The following pairs of lookout positions should have direct road connections constructed: 1 and 2, 1 and 5, 2 and
3, 3 and 4.
8 The greatest number of vehicles that could arrive at Z per hour, having come from A, is 2700.
ISBN 9780170395069 Answers 197

1 Firm A to do load 2. 2 Firm A to do load 1.

Firm B to do load 1. Firm B to do load 3.
(Total cost $240.) Firm C to do load 2.
(Total cost $340.)
3 Firm A to do load 2. 4 Operator A on machine 1.
Firm B to do load 4. Operator B on machine 2.
Firm C to do load 3. (Total number of components 180.)
Firm D to do load 1.
(Total cost $3100.)
5 Operator A on machine 2. 6 Operator A on machine 2.
Operator B on machine 1. Operator B on machine 1.
Operator C on machine 3. Operator C on machine 4.
(Total number of components 610.) Operator D on machine 3.
(Total number of components 375.)
1 Company A to task 2. 2 Company A to task 2.

Company B to task 1. Total cost 177 units. Company B to task 1.
Company C to task 3. Total cost 300 units.
3 Company A to task 3. 4 Company A to task 4.
Company B to task 1. Company B to task 5.
Company C to task 4. Company C to task 2.
Company D to task 2. Total cost 43 units. Company D to task 6.
Company E to task 1.
Company F to task 3. Total cost 135 units.
5 Company A to task 2. Benefit 120 units. 6 Company A to task 1. Benefit 29 units.
Company B to task 1. Benefit 240 units. Company B to task 3. Benefit 28 units.
Company C to task 2. Benefit 33 units.
or
Company A to task 3. Benefit 31 units.
Company B to task 1. Benefit 26 units.
Company C to task 2. Benefit 33 units
7 Company A to task 2. Benefit 240 units. 8 Company A to task 2. Benefit 40 units.
Company B to task 3. Benefit 140 units. Company B to task 1. Benefit 51 units.
Company C to task 4. Benefit 610 units. Company C to task 3. Benefit 42 units.
Company D to task 1. Benefit 160 units.
9 Company A to task 3. Benefit 33 units.
Company B to task 4. Benefit 43 units.
Company C to task 1. Benefit 35 units.
Company D to task 2. Benefit 45 units.
10 b Steps of algorithm not shown here. Optimal solution is as follows:
Adelaide rep oversees Sydney office. (Return airfare $380.)
Brisbane rep oversees Darwin office. (Return airfare $570.)
Hobart rep oversees Melbourne office. (Return airfare $290.)
Perth rep oversees Broome office. (Return airfare $720.)
(Total of return airfares is then $1960.)

11 a Quick Courier Co. to job 1. ($290) or Quick Courier Co. to job 4. ($150)
Speedy Courier Co. to job 3. ($260) Speedy Courier Co. to job 3. ($260)
Deliverit Couriers to job 2. ($80) Deliverit Couriers to job 2. ($80)
Competitive Couriers to job 4. ($140) Competitive Couriers to job 1. ($280)
Total cost of this minimum total cost solution is $770. Total cost of this minimum total cost solution is $770.
b If all four companies add $50 for job 4 the optimum allocation would not change, i.e. the two alternatives would still
be as above (but the total cost would increase to $820).
c If all four companies multiplied their prices for job 2 by 4 the optimum situation would change.
Quick Courier Co. would do job 1.
Speedy Courier Co. would still do job 3.
Deliverit Couriers would do job 4.
Competitive Couriers would do 2.
(Total cost of this minimum total cost solution now $970.)
d Competitive couriers. (Total $730.)
12 Jack to area 1. Jim to area 4. Judy to area 3.
Nicci to area 5. Ti to area 2. Total sales 50.
13 Ashok to location 3. Daniel to location 1. Kirsten to location 2. Shani to location 4.
14 Two possibilities:
A → 1 (40 km) B → 5 (30 km) C → 2 (30 km) D → 3 (40 km) E → 4 (40 km).
Or
A → 3 (30 km) B → 5 (30 km) C → 2 (30 km) D → 1 (50 km) E → 4 (40 km).
(Each possibility gives a total distance travelled of 180 km.)
15 Factory A to manufacture the Charger.
Factory B to manufacture the Exeon.
Factory C to manufacture the Bijou.
Factory D to manufacture the Devine.
Factory E to manufacture the Arcain.
Exercise 8C PAGE 160
1 Assign Alex to task 3 (15 minutes)

Assign Ben to task 4 (17 minutes)
Assign Theo to task 1 (21 minutes)
Working simultaneously all three will have finished after 21 minutes.
Yes, if all working together they could finish three tasks in 20 minutes if Alex assigned to task 4 (20 minutes), Ben to
task 1 (18 minutes) and Theo to task 3 (18 minutes). Working simultaneously the three tasks would be completed after
20 minutes.
2 Assign the digger from the Horton storage facility to Abbysix. (25 km)
Assign the digger from the Peel storage facility to Crampon. (18 km)
Assign the digger from Wimsley storage facility to Denton. (31 km)
Total distance 74 km.
3 Assign taxi A to Tenielle, who will have to wait 8 minutes.
Assign taxi C to Zarani, who will have to wait 6 minutes.
Assign taxi D to Bill who will have to wait 8 minutes.
4 Assign person B to job 2. ($220)
Assign person C to job 1. ($120)
Assign person D to job 3. ($160)
Assign person E to job 4. ($150)
Total profit is $650.
ISBN 9780170395069 Answers 199

Challenge PAGE 162
Optimum solution is Worker A to task 9. 15 minutes.

Worker B to task 6. 15 minutes.
Worker C to task 4. 21 minutes.
Worker D to task 7. 29 minutes.
Worker E to task 3. 9 minutes.
Worker F to task 10 15 minutes
Worker G to task 1. 17 minutes.
Worker H to task 2. 20 minutes.
Worker I to task 5. 14 minutes.
Worker J to task 8. 36 minutes. Total 191 minutes.
Miscellaneous exercise eight PAGE 163

Annual 10% 10%
Quarterly 10% 10.38% (2dp)
Monthly 10% 10.47% (2dp)
Weekly 10% 10.51% (2dp)
Daily 10% 10.52% (2dp)
2 a $553 382
b $553 382
3 Correct answer is $662 642, to the nearest dollar.
4 Immediately after the fifteenth repayment has been made the amount owed is $4908.28 (to the nearest cent).
5 27 hours, B - E - J - M
6 a The greatest number of messages that can go from A to G in an hour is 850.
b The greatest number of messages that can go from A to D in an hour is 650.
7 The maximum flow from A to X is 1200 units.
The extra capacity along EX increases the maximum flow from A to X by 150 units.
8 Fran should swim backstroke (54.2 seconds).
Marta should swim breaststroke (48.2 seconds).
Julie should swim butterfly (43.6 seconds).
Micah should swim freestyle (42.4 seconds).
Total time 188.4 seconds.

9 1 000000
900000
800000
700000
600000
500000
400000
300000
Raw data
200000
100000
0
1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014
Compare your comments describing any trends in the numbers of live cattle exported from Australia per year over the
period of time from 1988 to 2013 with those of others in your class.
10 Vehicle A to location 3 7 minutes.
Vehicle B to location 2 9 minutes.
Vehicle C to location 1 6 minutes.
Vehicle D to location 5 9 minutes.
Vehicle E to location 4 10 minutes.
11 a 6850
b 1st Quarter 4932
2nd Quarter 6028
3rd Quarter 7261
4th Quarter 9179
12 The total length of the minimum spanning D
H 17
tree for the given network is 100 units. 15 J
The minimum span is shown on the right. 18
B F
11 I
10 9
A
7 8 G
C 5 E
ISBN 9780170395069 Answers 201

13 a 25 days, A - C - F - G - J b 3 days
14 (Answers may vary slightly dependent on level of rounding used when calculating.)
a 1st two months index 0.3579 2nd two months index 0.5675
3rd two months index 1.6293 4th two months index 1.8760
5th two months index 1.1878 6th two months index 0.3815
b 1st two 2nd two 3rd two 4th two 5th two 6th two
months months months months months months
Year 1 3460 3190 3120 3370 3180 2760
Year 2 3860 3870 3940 3810 4100 4280
Year 3 4200 4560 4580 4380 4310 4690
c Viewing the graph indicates linear modelling is suitable. (It could be argued that the sixth deseasonalised value is
an outlier and hence exclude it from the calculation of the regression line but answers given here are based on the
inclusion of all points.)
d D = 95.52n + 2960
e i 1st two months of year 4, n = 19, estimatedN = 1710 (nearest ten).
ii 4th two months of year 4, n = 22, estimated N = 9500 (nearest ten).
f Compare your possible events with those of others in your class.
15 a 21 weeks, A - D - H - I - J - L b 11 weeks c 19 weeks
16 $256 000
F
E
I
H
B $256 000
A C J
D
G
Direct connection from H to I is worth doing if it can be done for less than $38 000 as it could then replace EC in the
minimum span.
17 Either assign Machine A to product 3 (3300 units)
Machine B to product 4 (4100 units)
Machine C to product 1 (4000 units).
or assign Machine A to product 4 (3300 units)
Machine B to product 3 (4100 units)
Machine C to product 1 (4000 units).
(Either way gives the total number of units produced as 11 400.)
18 a 1155 m b Goods In - Dock 1, Customs - Goods Out, Customs - Waiting,
Customs - Dock 3, Dock 1 - Dock 2, Dock 2 - Customs.
19 Allocate DG Transport to job 4. ($16 500)
Allocate Haulage and Co. to job 2. ($12 800)
Allocate JA Movers to job 1. ($18 000)
Allocate Move it and Shift it to job 3. ($25 000)
Total cost $72 300.

INDEX
common difference x
A common ratio x
compound interest x, 54–5
depreciation 64–5
algorithms xviii
effective annual interest rate 58–9
Hungarian 150, 151 initial deposit plus regular investments 60–2
Kruskal’s 95
loans 64, 65–7
Prim’s 94–5, 99–101
correlation xiv
for solving allocation problems 146–52
correlation coefficient xv
allocation problems 141–3, 162 cost matrix (allocation problems) 146–50
algorithm for solving 146–52
critical path 122–4
allocating for a maximum 143–4, 151–2
float or slack 122–3
allocating for a minimum 142–3, 144, 146–50 cuts, maximum flow 115
cost matrix 146–50
four machines, four tasks 144
number of people ≠ the number of tasks 157–9
D
square matrix 157–9
amortization calculators 71
annuities 79–81 dependent variable xiii
frequency of payments ≠ frequency of compounding 84–6 depreciation 64–5
indexed 80 deseasonalised data 39–43
arcs xi, 111 directed edges xi
arithmetic sequences (arithmetic progressions) x drawing down the investment
assignment problems 141–162 annuities 79–81
association between variables xiii, xv perpetuities 81
average percentage method 34–5 superannuation 77–9
B E
backward scan (project networks) 122 earliest start time 122, 123
bipartite graphs xii, xiii, 142 edges xi
bivariate data xiii effective annual interest rate 58–9
borrowings 64–7 explanatory variable xiii
extrapolation xv, 4
C
F
calculators
annuities 79, 84–6 finance, and moving averages 32
compound interest 55, 60–2, 66–7 fixed term annuities 79
geometric sequences xvii–xviii flat rate method 64
superannuation 78 float 122, 123
centred moving averages 28–9 forward scan (project networks) 122
coefficient of determination xv 4 point moving average 22, 28
ISBN 9780170395069 Index 203

minimum, allocating for a 142–3, 144, 146–50
G minimum completion time 122, 123
minimum cut 115
minimum spanning trees 93
geometric sequences (geometric progressions) x, xvii–xviii from distances table 99–101
graphs see networks Kruskal’s algorithm 95
Prim’s algorithm 94–5, 99–101
systematic approach 94–5
H moving averages 20–2, 26
centred 28–9
and finance 32
housing loans, affordability 70–1
seasonal effects 33–5
Hungarian algorithm 150, 151
and seasonally adjusted data 39–43
use of spreadsheets 27
I
N
increasing trend 5
independent variable xiii
networks xi–xiii
initial deposit plus regular investments 60–2
maximum flow 109–12, 115
interpolation xv, 4
minimum spanning trees 93–5, 99–101
investments
project networks 121–4
compound interest 54–5, 58–9, 60–2, 88
nominal annual interest rate 58
simple interest 53–4
number vii
see also annuities; perpetuities; superannuation
P
K
Pearson’s correlation coefficient xv

Kruskal’s algorithm 95
percentages vii–viii
perpetuities 81
positive linear correlation xiv
L
Prim’s algorithm 94–5, 99–101
project networks 121–4
latest start time 122, 123 constructing 129–30
life annuities 79 critical path 122–4
line of best fit xiv earliest start time 122, 123
line graphs 5 latest start time 122, 123
linear equations ix minimum completion time 122, 123
linear regression xiv, 7 systematic approach 122–4
linear relationships viii
loans 64
with regular payments 65–7 R
what price house can people afford? 70–1
real interest rate 60
recursion x–xi, xvii–xviii
M
recursive rules x, xi
annuities 80, 84–6
making predictions for simple and compound interest 55, 66–7
seasonally adjusted data 40–3 superannuation 78
time series data 6–10 reducing balance method 64
maximum, allocating for a 143–4, 151–2 regression xiv–xv, 7
maximum flow 109 response variable xiii
equals the minimum cut 115
systematic approach 110–12, 115

S T
scattergraphs xiv technology use xvi–xviii

seasonal effects 9, 20 3 point moving average 20–1
quantifying 33–5 time series data 3–6, 14–15
seasonal indices 33–5 general smoothing 26
seasonally adjusted data 39–40 making predictions 6–10
making predictions 40–3 seasonal variation 9
simple interest x, 53–4, 56 total float 123
simultaneous equations ix trees xii
sink (maximum flow) 109, 110 spanning 93
slack 122, 123 trend line xiv
smoothing data 20–2, 26 making underlying trends more apparent 20
solving equations ix
source (maximum flow) 109, 110
spanning trees 93–5, 99–101 U
spreadsheets
compound interest 60–2, 66
geometric sequences xvii underlying trends 20
moving averages 27 unit cost method 64
square matrix (allocation problems) 157–9
statistical investigation process xviii, 4
straight line graphs viii V
superannuation 77–9
accumulation phase 77 vertices xi
pension phase 77
whole life annuities 79
ISBN 9780170395069 Index 205

Sadler Apps Unit 4

Uploaded by

Copyright:

Available Formats

Sadler Apps Unit 4

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sadler Apps Unit 4

Uploaded by

Copyright:

Available Formats

Australia • Brazil • Japan • Korea • Mexico • Singapore • Spain • United Kingdom • United States

Mathematics Applications Unit 4 © 2017 A J Sadler

For product information and technology assistance,

National Library of Australia Cataloguing-in-Publication Data

1st revised edition

Mathematics--Study and teaching (Secondary)--Australia.

Cengage Learning Australia

Cengage Learning New Zealand

For learning solutions, visit cengage.com.au

Printed in China by China Translation & Printing Services.

ISBN 9780170395069 iii

iv MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

FINANCE II: DRAWING MAXIMUM FLOW 108 ASSIGNMENT

Spanning trees ............................ 93

ISBN 9780170395069 Contents v

vi MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

Typical percentage questions:

ISBN 9780170395069 Preliminary work vii

III Increase $40 by 28%.

IV Decrease $40 by 28%

IV 28% of an amount is $40. Find the amount.

Straight line graphs

viii MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

x = 3.46 (correct to 2 decimal places)

ISBN 9780170395069 Preliminary work ix

x MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

• The connections are called edges. 5 edges 2 4 5

Graphs may be directed or undirected.

A directed graph An undirected graph

ISBN 9780170395069 Preliminary work xi

The following are not trees:

Contains a cycle. Not connected. Contains a cycle.

xii MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

Rosemary Chris Taylor

Mid A&E ICU

ISBN 9780170395069 Preliminary work xiii

)noitalupop fo 000 001 rep raey reP(

esaesid traeh ot eud shtaeD

The location of the points suggests that as one of the

could be modelled by drawing a line of best fit or

With one variable generally increasing as the other y

The line of best fit allows predicted values to 200

Rather than having to draw this ‘line of best fit’

xiv MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

ISBN 9780170395069 Preliminary work xv

xvi MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

ISBN 9780170395069 Preliminary work xvii

I.e. define the sequence as: bn+1 =

The statistical investigation process

xviii MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

draW kcirtaP/otohP kcotS ymalA

Year of valuation 1948 1968 1988 2008 2015

ISBN 9780170395069 1. Time series data 3

(2) Plan the collection and use of appropriate data.

(3) Apply appropriate techniques to analyse the data.

4 MATHEMATICS APPLICATIONS Unit 4 ISBN 9780170395069

This particular line graph involves $200 000

Note that when graphing time

ISBN 9780170395069 1. Time series data 5

) m/stinu( level noitulloP