Foundations of Quantitative Methods

Foundations of
Quantitative Methods
FedUni
#52, Nagarjuna Hills, Hyderabad – 500 082
© FedUni, March 2005. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system, used in
a spreadsheet, or transmitted in any form or by any means – electronic,
mechanical, photocopying or otherwise – without prior permission in writing from
The Federation of Universities (FedUni)
Ref. No. FQM/A – 03 2K5 45
For any clarification regarding this book, please write to FedUni giving the reference number of the book, and the
page number.
While every possible care has been taken in preparing the book, FedUni welcomes suggestions from students for
improvement in future editions.
Contents
Chapter 1 Quantitative Methods: An Overview 1
Chapter 2 Basics of Mathematics 5
Chapter 3 Calculus 35
Chapter 4 Arranging Data 47
Chapter 5 Measures of Central Tendency 61
Chapter 6 Measures of Dispersion 79
Chapter 7 Index Numbers 93
Chapter 8 Introduction to Probability and Probability Distributions 106
Chapter 9 Matrices 126
Chapter 10 Interpolation and Extrapolation 148
Appendix 156
Introduction to Quantitative Methods
Detailed Contents
Chapter 1: Quantitative Methods - An Overview: Definition of Quantitative
Methods, Evolution of Quantitative Methods, Classification of Quantitative Methods,
Methodology of Quantitative Methods, Advantages of Quantitative Methods,
Limitations of Quantitative Methods
Chapter 2: Basics of Mathematics: Number Systems, Permutations and
Combinations, Logarithms, Progressions, Functions, Solving Simultaneous Equations
Chapter 3: Calculus: Limits, Derivatives: First order and Higher Order Derivatives,
Rules of Differentiation, Partial Differentiation, Optimization, Integral Calculus:
Indefinite integral
Chapter 4: Arranging Data: Meaning of Data, Types of Data, Data Collection,
Tables and Graphs as Data Presentation Devices, Frequency Distribution
Chapter 5: Measures of Central Tendency: Objectives of Averaging, Requisites of
a Good Average, Types of Averages, Mathematical Averages, The Median, The Mode
Chapter 6: Measures of Dispersion: Range, Inter Quartile Range and Quartile
Deviation, Mean Deviation, Variance, Standard Deviation, Bienayme Chebyshev’s
Rule
Chapter 7: Index Numbers: Definition of Index Number, Characteristics of Index
Numbers, Uses of Index Numbers, Types of Index Numbers, Problems Related to
Index Numbers, Methods of Constructing Index Numbers, Quantity and Value Indices
Chapter 8: Introduction to Probability and Probability Distribution: Basic
probability Concepts, Types of Probability, Probability Rules, Random Variables,
Types of Probability Distributions, The Normal Distribution
Chapter 9: Matrices: Definition of Matrix, Types of Matrices, Operations on
Matrices, Determinant of a Matrix , Properties of Determinants, Adjoint of a Matrix ,
Inverse of a Matrix , Application of Matrix Methods to Solve Linear Equations
Chapter 10: Interpolation and Extrapolation: Assumptions of Interpolation and
Extrapolation, Applications of Interpolation and Extrapolation, Methods of
Interpolation and Extrapolation
Chapter 1
Quantitative Methods: An
Overview
In this chapter we will discuss:
Evolution of Quantitative Methods
Classification of Quantitative Methods
Advantages of Quantitative Methods
Limitations of Quantitative Methods
Foundations of Quantitative Methods
Managers need to analyze environmental factors on a continuous basis, determine the

objectives, identify the best option from the set of available alternatives, implement,
coordinate, evaluate, and control the factors continuously to achieve set objectives. As
the environment has become complex, decision-making in organizations has also
become increasingly complex.
In this complex business environment, managers do not take decisions on the basis of
subjective factors like their experience, observation, or evaluation anymore. Decisions
are made after analyses of data that reveal relationships between them, indicate trends,
and show rates of change in the relevant variables.
Quantitative methods help managers collect, present, analyze, and interpret the
available data meaningfully. Managers can accomplish their predetermined objectives
like profit maximization, cost minimization, or efficient and effective use of production
capacities by using quantitative methods.
Quantitative methods are statistical and programming techniques, which help decision-
makers solve business problems. In other words, quantitative methods are the
techniques that provide decision-makers systematic and powerful means of analysis, on
the basis of quantitative data, for achieving predetermined objectives.
EVOLUTION OF QUANTITATIVE METHODS

The utility of quantitative methods has been realized long ago and the science of
mathematics is probably as old as the human society. However, with the evolution of
industrial engineering, scientific methodologies that were prominent earlier in the
natural sciences were found applicable to management functions - planning,
organizing, and controlling of operations.
In the late nineteenth century, Frederick W. Taylor proposed an application of a
scientific method to an operations management problem, namely, the productivity of
men shoveling ore. Prior to this development, there was a strong belief that the largest
shovel a worker could fill and carry was the shovel size which maximized the output.
Taylor questioned the validity of the assumption and performed a series of experiments
to disprove it. He determined that the only variable that was significant was the
combined weight of the shovel and its load. Too much weight on the shovel would
result in the worker getting tired soon and moving slowly, while too little load results
in too many trips. The experiments concluded that productivity could be raised
substantially by loading the shovel properly.
Another significant contribution to the evolution of quantitative methods was made by
Henry L. Gantt, who devised a chart, known as the Gantt chart, to schedule production
activities. Prior to his work, production was a haphazard exercise. The jobs processed
through one stage of production used to wait for days for acceptance in the next
processing center. The Gantt chart scheduled each job from one machine to another,
and minimized the delays in between.
CLASSIFICATION OF QUANTITATIVE METHODS

The wide variety of quantitative methods that are available in the modern times can be
broadly classified into two groups:
Statistical techniques
Programming techniques
Statistical techniques are used in conducting the statistical inquiry concerning a certain
phenomenon. They include various methods ranging from the collection of data to the
task of interpreting the collected data. The methods of collecting, classifying and
tabulating statistical data, the calculation of various statistical measures like mean,
2
Quantitative Methods: An Overview
standard deviation, coefficient of correlation etc., the methodologies of analyzing and

interpreting data, and finally the task of deriving inferences and judging their reliability
are examples of statistical techniques.
Programming techniques (also known as operations research techniques) are used by
many decision makers in modern times. They were first designed to tackle defense and
military problems and are now being used to solve business problems. They include a
variety of techniques like linear programming, games theory, simulation, network
analysis, queuing theory, and so on. These techniques involve the building of
mathematical models that relate the relevant variables in a situation to the outcome,
and yield solutions to problems in terms of the values of the variables involved. The
following sections describe the steps involved in the application of the programming
techniques.
ADVANTAGES OF QUANTITATIVE METHODS

Quantitative methods have been gaining significance since World War II in the areas
of business administration. They help managers tackle intricate and complex problems
in businesses. The methods are applied to achieve organizational targets that are as
diverse as selecting or promoting personnel, using materials efficiently, projecting
long-term capital requirements, forecasting demand, estimating customers’
preferences, etc. These methods replace the intuitive and subjective approach to
decision-making with an analytical and objective approach.
Quantitative methods are used in the fields of procurement, production, marketing,
finance and other related fields. They can be used to solve problems like the allocation
of available resources to various products so that, either the profits are maximized or
costs are minimized within a specified time frame. They can also be used by operations
managers to determine the number of men and machines to be employed and used at
any point of time to ensure that neither of them remain idle and customers are
satisfactorily serviced.
At times, in a business, there can be confusion in finding an acceptable definition of
the long-range objectives. Managers may face a dilemma in making a choice between
profit maximization and cost minimization as the long-term objective. Quantitative
methods (like linear programming) can help in solving such problems as they permit
systematic evaluation of different strategies for attaining different objectives. The
following sections explain the role of quantitative methods in various areas of decision-
making:
1. Quantitative methods like regression analysis help managers predict the unknown
value of a variable from the known variables. This method can be used to predict
sales for a period on the basis of past data. The cost of production can be estimated
for a particular production level.
2. Quantitative Methods like hypothesis testing are used to test the significance of a
population parameter on the basis of sample. Practically, it is not possible to
estimate a parameter by surveying the entire population, and hence managers rely
mainly on the sample data and then check whether the sample data represents the
population.
3. Quantitative methods like decision theory help managers make decisions under
uncertainty taking various courses of action into consideration. Decision theory
can be used by managers to decide whether to outsource a function. The method
also gives the outcome of a decision in monetary terms.
4. Quantitative methods like linear programming help managers allocate scarce
resources in an optimum manner while solving the problems involved in
scheduling, product mix determination, and so on.
3
Apart from the above advantages, quantitative methods like probability, variance
analysis, time series analysis, and index numbers are other efficient tools used by
managers to solve their organizational problems. These methods are discussed in
greater detail in the later chapters.
LIMITATIONS OF QUANTITATIVE METHODS
Though quantitative methods are applied to an increasingly large number of business

problems, they have certain limitations. The following are some of the important
limitations of quantitative methods:
1. Quantitative methods involve the use of mathematical models, equations and other
mathematical expressions, which are derived on the basis of several assumptions.
Such underlying assumptions that are present in the current problem may or may
not be relevant to some other problems. If this caution is not taken into
consideration, wrong applications of quantitative methods may yield disastrous
solutions.
2. Quantitative methods are usually expensive as they require the services of
specialized people. Even big organizations use quantitative methods to a limited
extent because many applications are not worth their costs. Instead of using
quantitative methods, managers generally prefer to exercise their intuition and
judgment to make decisions.
3. The accuracy of the solution attained by quantitative methods is also hindered by
pitfalls like inadequacy of data, inconsistency in definitions, selection of the wrong
sample, improper choice of the method, inappropriate comparisons, and improper
presentations.
4. Quantitative methods cannot be used to study qualitative phenomena as they do
not take the intangible and non-measurable human factors into consideration. For
instance, the methods make no allowances for intangible factors such as skill,
attitude, and vigor of managers in taking decisions. However, the methods can be
successfully applied indirectly by first converting intangible expressions into
quantitative terms. For example, the intelligence of a manager can be quantified by
providing different scores to difference qualifications.
SUMMARY
As the process of decision-making is becoming increasingly difficult, managers can no

longer afford to make decisions on the basis of subjective factors like experience,
observation, or evaluation alone. They have to use objective and quantitative methods
to collect, present, analyze and meaningfully interpret the available data to arrive at
proper solutions.
Quantitative methods involve the use of numbers, symbols, mathematical expressions,
and other quantitative elements and serve as supplements to the subjective intuition of
the decision -makers. The utility of these methods has been realized from long ago.
This chapter discussed the historical evolution of quantitative methods and the
classification of quantitative methods. The chapter also discussed the advantages and
limitations of quantitative methods.
4
Chapter 2
Basics of Mathematics
• Number Systems
• Permutations and Combinations
• Logarithms
• Progressions
• Functions
• Simultaneous Equations
Before introducing new concepts of quantitative methods, let us revisit a few basic
concepts in Mathematics. The decimal, binary and octal number systems along with the
conversions are discussed in the chapter. The concepts of permutations and
combinations are used while calculating probability of a random event. The concept of
progression is used to understand the trend of a data series. The three types of
progression: arithmetic, geometric and harmonic progression is discussed in detail.
The relationship between the dependent and independent variable can be linear,
exponential, logarithmic or inverse. The concept of functions discusses these
relationships in detail.
NUMBER SYSTEMS
The two number systems that we generally come across are decimal system (0, 1, 2, 3,
4, 5, 6, 7, 8, 9) and Roman system (I, II, III, IV, V...). But there are several other
number systems such as binary, octal, hexadecimal etc. The binary number system,
which contains only two digits 0 and 1, is widely used in electronic devices. The two
states of an electronic device ‘ON’ or ‘OFF’ are denoted by 1 and 0 respectively. John
Von Neumann suggested in 1946 that a computer should also use this kind of two way
binary system. Such a system is known as John Von Neumann architecture and most of
the computers use this architecture. Similar to the case of an electronic device when the
switch is said to be ‘ON’ or ‘OFF’, a computer uses two signals - pulse and no pulse.
Pulse is denoted by the digit ‘1’ and no pulse is denoted by the digit ‘0’. The other
number systems are octal number system based on the digits 0 to 7, the hexadecimal
number system based on the digits 0 to 9 and the letters A to F. The decimal system is
the most commonly used number system.
Decimal System
The decimal system comprises of ten basic symbols or digits (0, 1, 2, 3, 4, 5, 6, 7, 8
and 9) and is based on ‘tens’ i.e., each of the digit is a number and has a value or
importance ten times greater than the digit just to its right.
For example, 9875 means reading from the right
5 lots of 1(units) = 05
7 lots of 10 (tens) = 70
8 lots of 100 (hundreds) = 800
9 lots of 1000 (thousands) = 9000
Total = 9875
Each symbol in decimal system has its own absolute value and place value. In the
above example, 9 have got two values, i.e., one its absolute value which is 9 and its
place value which is 9000. Similarly the numbers 8, 7 and 5 have two values; 8, 7 and
5 as absolute values and 800, 70 and 5 as place values respectively. Number 10, which
is called the base or decimal system, is not a basic digit in this system. It is a result of
two digits 1 and 0 placed in order after the last digit of number system i.e. 9.
Thus, 10 = 1 ten + 0 unit
Similarly,
086 = 0 hundreds + 8 tens + 6 units
864 = 8 hundreds + 6 tens + 4 units
As the decimal system has number 10 as its base or radix, the position of the digit in
this system (base to system) determines the magnitude of the numbers. For example the
numbers 86 and 864 can be expressed in the base to system as
6
086 = 0 x 102 + 8 x 101 + 6 x 100

864 = 8 x 102 + 6 x 101+ 4 x 100
The base of a number can be defined as the number which when raised to zero power
assigns the lowest position value, raised to the power of one assigns the second
position value, raised to the power of two assigns the third position value and so on.
A number in any number system can be expressed in general term as
N = dn+1 × rn +.........d3r2 +d2 r1 +d1 r0 (2.1)
Where, N = any number of any base system
dn = digit in the nth position
rn = the exponent for (n+1)th digit position.
Example 2.1
‘543’ is a number in decimal number system. What are the place value and the absolute
value of the digit ‘4’?
Solution:
Each symbol in decimal system has its own absolute value and place value. The
absolute value is 4 and the place value is 4×101 = 40
The Binary Number System

In the binary system, there are only two digits (0 and 1) and the system is based on
only two variables. In this number system, each of the digits in a number is twice the
value of the digits on its right. For example 1001 means, reading from right to left:
1 lot of 1 = 1
0 lot of 2 = 0
0 lot of 4 = 0
1 lot of 8 = 8
Total = 9
Similarly, the binary number 11111 will mean
1 lot of 1 = 1
1 lot of 2 = 2
1 lot of 4 = 4
1 lot of 8 = 8
1 lot of 16 = 16
Total = 31
As the binary system has number 2 as its base or radix, the value of the position in the
binary system varies as powers of the two, i.e. as units, twos, fours, eights, sixteens,
thirty twos and so on.
For example, the binary numbers 1001 and 1111 can be expressed in decimal system as
(1001)2 = 1×23 + 0×22 + 0×21 + 1×20 = 8 + 0 + 0 + 1 = (9)10
(1111)2 = 1×23 + 1×22 + 1×21 + 1×20 = 8 + 4 + 2 + 1 = (15)10
In this system a carry over occurs every time two is reached similar to the carryover
that occurs when 10 is reached in the decimal system.
7
Example for carryover in Decimal system

29
+ 31
Carry over 1
Total 60
Example for carryover in Binary system
1001
+ 1111
Carry over 111
Total 11000
In general, any number having ‘n’ bits in binary representation can be expressed as
P1 × 2n-1 + P2 × 2n-2 + P3 × 2n-3 + P4 ×2n-4+ ..........+ Pn-1 × 21 + Pn ×20
where P1 and Pn are the numbers on the extreme left and the numbers on the extreme
right respectively. Then two extreme numbers P1 and Pn are also called as most
significant bit (MSB) and least significant bit (LSB) respectively.
Conversion from decimal to binary
To convert a number from decimal system to binary system, divide the decimal number
by two until it is reduced to zero. In this process when an odd number is divided by
two, it leaves a remainder of 1, which is binary number 1; when an even number is
divided by two, it leaves a remainder zero, which is binary number zero.
Example 2.2
Convert the decimal system number 25 to binary system.
Solution:
25 ÷ 2 = 12, remainder 1
12 ÷ 2 = 6, remainder 0
6 ÷ 2 = 3, remainder 0
So the binary equivalent is = 11001
The binary number’s MSB is the last remainder or remainder at the bottom and the
LSB is the first or top remainder.
Conversion from binary to decimal
A number in the binary system can be converted into decimal system by multiplying
binary place value with a binary digit.
Example 2.3
Convert the following binary numbers into their decimal equivalents.
(a) 111001 (b) 111000 (c) 1100101
Solution:
Method 1
(a) (111001)2 = 1×25 + 1×24 + 1×23 + 0×22 + 0×21 + 1×20
= 32 + 16 + 8 + 0 + 0 + 1
= (57)10
8
(b) (111000)2 = 1×25 + 1×24 +1×23 + 0×22 + 0×21 + 0×20

= 32 + 16 + 8 + 0 + 0 + 0
= (56)10
(c) (1100101)2 = 1×26 + 1×25 + 0×24 + 0×23 + 1×22 + 0×21 + 1×20

= 64 + 32 + 0 + 0 + 4 + 0 + 1
= (101)10
Fractional Numbers
The discussion in this section would be on fractional numbers both in binary number
system and decimal number system.
A fraction such as 0.7 in the decimal number system is written as 7/10. This can also
be written as 7×10-1. Similarly 0.72 can be written as 72/100 = 72 × 10-2 = 7 × 10-1 + 2
×10-2.
The (.) dot is read as ‘decimal point’ in the decimal system and ‘binary point’ in binary
system.
Conversion of fractional binary to decimal number
Convert the number (0.110101)2 to its decimal equivalent.
(0.110101)2 = 1×2-1 + 1×2-2 + 0×2-3 + 1×2-4 + 0×2-5 +1×2-6
= 0.5 + 0.25 + 0 + 0.0625 + 0 + 0.015625
= (0.828125)10
Convert the number (101.11)2 to its decimal equivalent.
(101.11)2 = (1×22 + 0×21 +1×20) . (1×2-1+1×2-2)
= (4 + 0 + 1) . (0.5 + 0.25)
= (5.75)10
Conversion of fractional decimal to binary number

Example 2.4
Convert (0.671875)10 into its binary equivalent.
Solution:
0.671875 × 2
1 0.343750 × 2
0 0.687500 × 2
1 0.375000 × 2
0 0.750000 × 2
1 0.500000 × 2
1 0.000000
The number is obtained by writing the integer part of all remainder numbers (after
overflow over 2) one after another from left to right. Thus, the number is (0.101011)2.
Example 2.5
Convert (0.887651)10 into its nearest binary equivalent up to 6 digits.
9
Solution:
0.887651 × 2
1 0.775302 × 2
1 0.550604 × 2
1 0.101208 × 2
0 0.202416 × 2
0 0.404832 × 2
0 0.809664
The nearest equivalent is (0.111000)2
Let us verify and see
(0.111000)2 = 1×2-1 + 1×2-2 + 1×2-3 + 0×2-4 + 0×2-5 + 0×2-6
= 0.5 + 0.25 + 0.125 + 0 + 0 + 0
= 0.875 ≅ 0.887
The number (0.111000)2 is not exactly (0.887651)10 but is approaching it. If the
division is further carried out successively the number obtained may be approximately
the same as the original number. The number obtained is not exact because fractions
other than 1/2, 1/4, 1/8 ...1/2n cannot become zero after the binary point as in the above
example.
Example 2.6
Add the two binary numbers (111)2 and (101)2.
Solution:
Carry over (1) (1)
1 1 1
+ 1 0 1
11 0 0
(111)2 + (101)2 = (1100)2
Octal Number System

The base or the radix in an octal number system is 8 and it comprises the digits from 0
to 7 i.e. 0, 1, 2, 3, 4, 5, 6 and 7. Octal system is used in the computers for direct input –
output operations.
Conversion from decimal to octal
A number in the decimal system can be converted to octal system by successively
dividing the integral part of the number by 8.
This division is to be carried out till the integral part is exhausted and the remainders
will constitute the integral part of the octal number.
Example 2.7
Convert decimal 879 into its octal equivalent.
10
Solution:
Quotient Remainder
8 879 7
8 109 5
8 13 5
8 1 1
0
Thus the integral part of (879)10 = (1557)8
Conversion of octal into decimal
The decimal value of an octal number is equal to the sum of the decimal values of the
octal digits.
Example 2.8
Convert (357)8 to its decimal equivalent.
Solution:
3578 = 3 × 82 + 5 × 81 + 7 × 80
= 3 ×64 + 5×8 + 7×1
= 192 + 40 + 7
= (239)10
Conversion of fractional decimal to octal

The fractional part of the decimal number has to be multiplied by 8 successively,
nothing down the carry over which will form the fractional part of the number.
Example 2.9
Convert (0.32)10 into octal system.
Solution:
0.32×8
2 0.56 ×8
4 0.48×8
3 0.84×8
6 0.72×8
5 0.76×8
6 0.08 ×8
0 0.64x8
5 0.12
Thus, decimal 0.32 ≅ Octal 0.24365605
Conversion of fractional octal to decimal
Example 2.10
Convert (0.752)8 to its decimal system equivalent.
Solution:
(0.752)8 =7×8-1 + 5 × 8-2 × 2 ×8-3
= 7×0.125 +5×0.015625 + 2×0.001953125
11
= 0.875 + 0.078125 + 0.00390625

= (0.95703125)10
The integral parts and the fraction are determined separately similar to decimal &
binary conversions.
Example 2.11
Convert (65)8 to its binary equivalent.
Solution:
(65)8 = 6 × 81 + 5 × 80
= 48 + 5
= (53)10
53 ÷ 2 = 26, remainder 1
26 ÷ 2 = 13, remainder 0
13 ÷ 2 = 6, remainder 1
So the binary equivalent is = 110101
Alternatively, by converting each digit of octal number into its binary equivalent, we
can get binary equivalent of an octal number. There are two digits (6 and 5) in the octal
number. Binary equivalents of 6 and 5 are 110 and 101 respectively. Therefore, the
equivalent octal number is 110101.
PERMUTATIONS AND COMBINATIONS
Each of the arrangements that can be made by taking some or all of a number of
things/objects is called a permutation. Suppose there are four distinctive things w, x, y
and z. The permutations that can be made by taking two things at a time from the given
four things are each of these presenting a different arrangement of two letters.
wx wy wz xy xz xw
yz yw yx zw zx zy
Similarly, each of the groups or selections that can be made by taking some or all of a
number of things/objects is called a combination. Consider the same example as above.
The combinations that can be made by taking two things at a time from the given four
things are as below.
wx, wy, wz, xy, xz and yz
It is quite important to understand the difference between a permutation and a
combination. To form a combination we are concerned with the number and kind of
things/objects each group or selection contains; whereas to form a permutation we also
have to consider the order of the things that make up each arrangement.
Suppose there are four distinctive things w, x, y, z and a selection of three things out of
the given four at a time are
wxy, wyz, wxz and xyz
12
This is a combination. But for each combination above, we have six different
arrangements. The arrangements for the first combination are as follows:
wxy, wyx, xyw, xwy, ywx and yxw
Thus, permutation takes the order of things into consideration other than the number
and kind of things /objects.
Notation
The number of permutations of ‘n’ dissimilar thing taken ‘r’ at a time is denoted by nPr.
The number of combinations of ‘n’ dissimilar thing taken ‘r’ at a time is denoted by
n
Cr. Before going into the details of permutations and combinations, a function
principle is to be explained.
If one operation can be performed in ‘m’ different ways, (when it has been performed
in any one of these ways) and if another operation can then be performed in ‘n’
different ways then the two operations in succession can be performed in (m×n) ways.
Example 2.12
Find the numerical value of 10 P5 .
Solution:
n
Pr = n!
( n − r )!
10 10!
P5 =
(10 − 5)!
10 × 9 × 8 × 7 × 6 × 5!
=
5!
= 30240
Example 2.13
10
Find the numerical value of C5 .
Solution:
n n!
Cr =
r! (n − r )!
10 10!
C5 =
5!(10 − 5)!
10 × 9 × 8 × 7 × 6 × 5!
=
5!×5!
10 × 9 × 8 × 7 × 6
=
5 × 4 × 3 × 2 ×1
= 252
13
Example 2.14
Consider three persons (say A, B and C) coming to attend an interview in which there
are four panels (say 1, 2, 3 and 4). In how many ways can they be assigned to a panel
each at a different one?
Solution:
The first candidate, ‘A’ can be assigned to any of the four panels - 1, 2, 3, or 4. When
A has been assigned to any one of these panels, the second candidate, ‘B’ can be
assigned to any of the remaining three panels. The third candidate, ‘C’ can be assigned
to any of the remaining two panels. Thus, once A has been assigned in 4 ways, B has
been assigned in 3 ways and C in 2 ways, the total number of ways in which these
candidates can be assigned in 4×3×2 = 24.
Example 2.15
Consider 5 taxis running between Delhi and Agra. In how many ways can a person go
from Delhi to Agra and return by a different taxi?
Solution:
A person can go from Delhi to Agra in any of the 5 taxis (i.e. in five ways); and with
each of these there is a choice of four ways of returning (since the person is not to
come back by the same taxi); hence the number of ways of making the two journeys is
5×4 = 20 ways.
The number of permutations of n dissimilar things taken r at a time is given by,
n
Pr = n!
( n − r )!
n and r are positive integers and n ≥ r.

where n! is the continued product of the first n natural numbers. It is read as “factorial
n” and also denoted as ∠n . It is to be noted that 0! is defined as unity and 1! is also
equal to 1.
This principle may be extended to the case in which there are more than two operations
each of which can be performed in a given number of ways.
The number of permutations of ‘n’ things taken all at a time is
n
Pn = n!
=
n! n!
= = n!
( n − n )! 0! 1
n
Pr can also be calculated.
n
Pr = n(n-1Pr-1) = (n)(n-1) n-2Pr-2
Example 2.16
How many different numbers can be formed by using five out of the nine digits 1, 2, 3,
4, 5, 6, 7, 8 and 9?
Solution:
The number of dissimilar things = n = 9 and we have to find out the number of
permutations of them taken five at a time = r = 5.
n
Pr = 9P5 = 9! 9! 9 × 8 × 7 × 6 × 5 × 4! = 15120
= =
(9 − 5)! 4! 4!
The number of combinations on ‘n’ dissimilar things taken ‘r’ at a time is given by
n n!
Cr =
r! (n − r )!
14
Example 2.17
In how many ways can 6 balls be selected from a box containing 10 balls?
Solution:
The number of balls in the box=n=10.
We have to select 6 balls out of the 10. Hence r=6.
10! 10!
Number of ways of selecting 6 balls out of 10 =10C6= = =210.
(10 − 6)!×6! 4!×6!
Example 2.18
How many different numbers can be formed by using four out of the seven digits 1, 2,
3, 4, 5, 6, and 7?
Solution:
The number of dissimilar elements = n = 7 and we need to find out the number of
permutations if four of them are selected at a time = r = 4.
nP = 7 P4 = 7! 7! 7 × 6 × 5 × 4 × 3! = 840
r = =
(7 − 4)! 3! 3!
Example 2.19
n
P n +1 P7
If 6 : = 1:8, then find n.
Solution:
n n!
P6 =
(n − 6)!
n +1 (n + 1)!
P7 =
(n − 6)!
n
P6 n! (n − 6)!
n +1
= ×
P7 (n − 6)! (n + 1)!
n! 1
= =
(n + 1)n! (n + 1)
n
P6 1 1
n +1
= =
P7 8 (n + 1)
n +1 = 8
n=7
LOGARITHMS
The ‘logarithm’ of any number to a given base is the index or the power to which the
base must be raised in order to equal the given number. Thus if N, a (≠1) being any two
real numbers and for some real x if N= ax, then x is said to be the logarithm of N to the
base a, and is written as log aN= x. For example 34 = 81, the logarithm of 81 to base 3
is 4.
Functions defined by such equations are called logarithmic functions. The domain of
logarithmic functions is the set of positive real numbers and the range is the set of real
numbers.
If N = ax
log aN= log aax
15
loga N= x log aa
⇒ loga N = x (Since, loga a =1)
Example 2.20
Find the logarithm of 64 5 4 to base 4 2

Solution:
Let x be the required logarithm. Then by definition,
5
log 64 4
=x
4 2
(4 2 )x = 645 4
1 2
∴ (2 2
× 2 2 )x = 2 6
×2 5
5x 32
∴2 2 = 25
As bases are equal, indices can be equated
5x 32
∴ =
2 5
64
⇒ x= = 2.56
25
Example 2.21
4
Find the logarithm of 81 3 to the base 9 3 .
Solution:
Let x be the required logarithm; then by definition.
( 9 3 )x = 814 3
∴ (32. 31/2)x = 34. 31/4

5 17
x
∴ 32 = 3 4
As bases are equal, indices can be equated
5 17
∴ x= 1 x = 1.7
2 4
Any number can be taken as the base of logarithms, and corresponding to each base the
logarithm of a number can be found. But once it is understood that a particular base of
logarithms is in use, the base is omitted. For example, arithmetical calculations are
written as log 6, log 5 instead of writing it as log10 6and log10 5.
The logarithm of 1 to any base is zero, since any number to the power of zero equals 1
so log 1 to any base is zero.
The logarithm of any number (≠ 1) to itself as a base is always unity, since any number
to the power of one gives the number itself i.e. say a1 = a, so logaa =1.
16
To Find the Logarithm of a Product

Let (P × Q) be the product of two numbers P and Q where P>0 and Q>0.
Also suppose x = loga P and y = loga Q
⇒ ax = P and ay = Q
∴ P × Q =ax × ay
⇒ P × Q = ax + y
Hence by definition loga (P × Q) = x+y
Thus the logarithm of a product of two numbers is equal to the logarithm of the first
number plus the logarithm of the second number.
Also log (P × Q) = log P + log Q
To Find the Logarithm of a Fraction

Let (P/Q) be the fraction such that P > 0 and Q > 0.
Also suppose x = loga P and y = loga Q
⇒ ax = P and ay = Q
∴ P/Q =ax /ay
⇒ P/Q = ax - y
Hence by definition loga (P/Q) = x-y
Thus the logarithm of a fraction is equal to the logarithm of the numerator minus the
logarithm of the denominator.
Also log P/Q = log P – log Q
To Find the Logarithm of a Number Raised to Any Power, Integer or Fraction

Let loga (Pm) be required, and suppose
x = logaP, so that ax = P
∴ Pm = (ax)m
⇒ Pm = axm
By definition, log a Pm = xm
⇒ loga Pm = m. loga P
Thus, the logarithm of the mth power of a number is m times the logarithm of the
number.
1
m1
Similarly, log ap log ap
=
m
Thus, the logarithm of the mth root of a number is equal to 1/m times the logarithm of
the number.
Hence, the conclusion from the above three theorems is that by the use of logarithms,
the multiplication and division operations are replaced by those of addition and
subtraction respectively; and the operations of involution and evolution by those of
multiplication and division respectively.
Sign of the Logarithm

If (x > 1, a >1) or (0< x < 1, 0< a < 1) then logax > 0.
If (x > 1, 0<a <1) or (0< x < 1, a < 1) then logax < 0.
17
The following are inequalities in logarithms:

If a > 1, and x > a ⇒ logax > 1
If a > 1, and x < a ⇒ 0< logax < 1
If 0 < a < 1, and x > a, ⇒ 0 < logax < 1
If a >1, and x > y ⇒ logax > logay
If 0 < a < 1, and x > y ⇒ logax < logay
Example 2.22
3 9 81 2
Find the value of the expression is log − log + log + 3 log
4 4 8 3
Solution:
3 9 81 2
log − log + log + 3 log
4 4 8 3
2 4
3 3 3 2
= log − log + log + 3 log
2 2 3 3
2 2 2
= log 3 − 2 log 2 − 2 log3 + 2 log2 + 4 log 3 – 3 log 2 + 3 log 2 − 3 log 3

=0
PROGRESSIONS
Progressions are of three types
• Arithmetic Progression
• Geometric Progression
• Harmonic Progression
Arithmetic Progression
Quantities are said to be in arithmetic progression (A.P.) when they increase or
decrease by a common difference. The common difference is calculated by subtracting
any term of the series from its immediate succeeding term. The following are two
examples for an arithmetic progression:
4, 8,12,16,20...
-5,-10,-15,-20,-25...
In the above example, the first series is increasing by 4 i.e., a common difference of +4
and the second series is decreasing by 5 i.e., a common difference of -5.
The general form of an arithmetic progression is
a, (a+d), (a+2d), (a+3d), (a+4d), ........ [a+(n-1)d]+......
Where the first term is ‘a’ and the common difference is‘d’. Common difference is the
difference between any two consecutive terms.
It can be noticed that in any term, the coefficient of‘d’ is always less by one than the
number of term in the series. For example,
2nd term = a + d
3rd term = a + 2d
10th term = a + 9d
18
Therefore, the nth term in an A.P. is given by

Tn= a + (n-1) d
As the common difference is the difference of any term and its immediate previous
term, it can be expressed in the general form as
d = tn - tn-1 where tn is the nth term of the series.
Proof: d = [a + (n-1) d] - [a + {(n-1) -1}d]
= [a + nd-d] - [a + (n-2) d]
= a + nd-d-a-nd + 2d
=d
To find the sum of number of terms in arithmetical progression

Consider the general form of an arithmetic progression i.e., a, a+d, a+2d, a+3d,
a+4d,.........a+(n-1)d
Here ‘a’ is the first term of the series
‘d’ is the common difference and
a + (n-1) d is the nth term i.e. ‘Tn’.
Let ‘S’ denote the sum of the A.P. And ‘L’ the last term of the series.
∴ S = a + (a + d) + (a + 2d) + (a + 3d) +...+ [a + (n-1) d] +...+ (L-3d) + (L-2d) + (L- d)
+L
Writing the series in the reverse order
S = L+ (L-d) + (L-2d) + (L-3d) +...+ [a + (n-1) d] +...+ (a + 3d) + (a + 2d) + (a + d) + a
Now, adding both these series
⇒ 2S = (a + L + [a + d + L-d] + [a + 2d + L-2d] +.... to n terms
⇒ 2S = (a + L) + (a + L) + (a + L)... to n terms
= n (a + L)
n
⇒ S= (a + L)
2
∴ Sum to first ‘n’ terms of a series
n
= [first term + last term or nth term]
2
n n
= [a + l] or [a + {a + (n-1) d}]
2 2
n
= [2a + (n-1)d]
2
∴ Sum to n terms of a series,
n
Sn = [2a + (n-1) d]
2
Example 2.23
Find the sum of the series 1/4, 1/2, 3/4...to 20 terms.
Solution:
In the given series, the first term is ¼ and the common difference is 1/2 - 1/4 = 2/8 =
1/4
19
The sum to n terms of an A.P. is given by

n
Sn = [2a + (n-1) d]
2
S20 = 20[2(1/4) + (20-1)1/4] / 2
=10[1/2 +19(1/4)]
=10(1/2 +19/4)
=10(21/4) =105/2 = 52.5
Example 2.24
Find the sum of the series 20, 50, 80… to 25 terms
Solution:
Sum of first ‘n’ terms of a series is given by
Sn = (n/2) [2a + (n-1) d]
Where n = Number of terms in the series = 25
a = First term in the series = 20
d = Common difference = 30
Sn = (25/2) [2 × 20 + (25 – 1) 30]
= (25/2) [40 + 720]
= 12.5 × 760
= 9500
Example 2.25
Find the number of terms and the common difference of a series whose first term is 5,
last term 50 and the sum 275.
Solution:
Let ‘n’ be the number of terms in the series.
Given the first term of the series is 5 i.e. a = 5 and the last term is 50 i.e. Tn= 50
The sum of n terms of a series is given by
n
Sn = [first term + last term]
2
275 = n [5+50] / 2
⇒ 275 = n (55) / 2
⇒ n = 275×2 / 55
= 5×2 = 10
∴ The number of terms in the series is 10.
Let‘d’ denote the common difference of the series. Then the 10th term of the series
becomes a+9d (because the coefficient of a term is one less than the number of the
term in the series).
⇒ a + 9d = 50
⇒ 5+9d = 50
⇒ 9d = 50-5 = 45
⇒ d = 45/9 = 5
∴ The common difference of the series is 5.
20
Example 2.26
The sum of 20 numbers in arithmetic progression is 650. If the first term of the series is
4, what is the common difference between the terms?
Solution:
Sum of ‘n’ terms of an arithmetic progression is given by,
n[2a + (n − 1)d ]
Sn =
2
20[(2 × 4) + 19 × d ]
650 =
2
1300 = 20(8 + 19d)
1300
= 8 + 19d
20
(65-8)/19 = d
∴d = 3
Example 2.27
The sum of first 10 numbers in an arithmetic progression is 255. If the common
difference is 5, what is the first term of the series?
Solution:
Sum of ‘n’ terms of an arithmetic progression is given by,
n[2a + (n − 1)d ]
Sn =
2
10[2a + 9 × 5]
255 =
2
510 = 20a + 450
510 − 450
∴a =
20
a=3
Example 2.28
The 5th term of the arithmetic progression is 18 and the 10th term is 33. Find the 16th
term of the series.
Solution:
The nth term in an A.P. is given by
Tn= a + (n-1) d
Thus the 5th term is given by
18 = a + 4d --(A)
and the 10th term is given by
33 = a + 9d --(B)
Subtracting equation (B) from equation (A) gives
15 = -5d
∴d = 3
21
Substituting the value of d in equation (A), we get

18 = a + 4 × 3
18 = a +12
∴a = 6
Thus, the nth term for this series will be given by
Tn= 6 + (n-1) 3
Thus the 16th term is given by
T16 = 6 + (16-1) 3
= 51
Arithmetic mean
Suppose three numbers are in A.P. Then the middle term is said to be the arithmetic
mean of the other two.
If a-d, a, a + d are in A.P., then
Arithmetic Mean, A.M. is ‘a’
Suppose a, A, b are in A.P.
⇒ b - A = A-a [The common difference is equal]
⇒ b + a = 2A
⇒ A = (a + b)/2
∴ The Arithmetic mean, A = (a + b)/2
If ‘n’ arithmetic means are inserted between ‘a’ and ‘b’ such that the whole series thus
formed shall be in A.P., then there are (n + 2) terms including the first term ‘a’ and the
last term ‘b’.
⇒ b is the (n + 2)th term
⇒ b = a + (n + 1) d
⇒ d = (b - a) / (n +1)
Therefore the required means between ‘a’ and ‘b’ are
a + (b-a)/(n + 1), a + 2(b - a)/(n + 1), a + 3(b-a)/(n + 1)... and,
Kth A.M. = a + K (b - a) / (n+1)
Also sum of the ‘n’ arithmetic means between ‘a’ & ‘b’
n n
= (a + b) [since Sn = {first term + last term}]
2 2
Example 2.29
If the sum of five numbers in an A.P. is 35, and the sum of their squares is 285, find the
numbers.
Solution:
Let a - 2d, a - d, a, a + d, a + 2d be the five numbers in the series.
∴ (a - 2d) + (a - d) + (a) + (a + d) + (a + 2d) = 35
⇒ 5a = 35 ⇒ a = 7
22
Also given that

(a - 2d)2 + (a - d)2 + (a)2 + (a + d)2 + (a + 2d)2 = 285
⇒ [(a2 + 4d2- 4ad) + (a2 + d2 - 2ad) + a2 + (a2 + d2 + 2ad) + (a2 + 4d2 + 4ad)] = 285
⇒ a2 + 4d2 + a2 + d2 + a2 + a2 + d2 + a2 + 4d2 = 285
⇒ 5a2 +10d2 = 285
⇒ 5(a2 + 2d2) = 285
⇒ a2 + 2d2 = 285/5 = 57
⇒ (7)2 + 2d2 = 57
⇒ 2d2 = 57- 49 = 8
⇒ d2 = 8/2 = 4
⇒ d=2
∴ The numbers are [7 - 2(2)], [7-2], 7, [7+2], [7+2(2)]
i.e., 3, 5, 7, 9 and 11.
Example 2.30
Find the sum to first ‘r’ terms of an A.P. whose nth term is 2n + 1.
Solution:
The first term of the given A.P. = 2(1) + 1 = 2+1 = 3
The rth term of the given A.P. = 2(r) + 1 = 2r + 1
n
Sum to n terms is given by Sn = (first term + last term)
2
= (r/2) × (3 + 2r + 1)
= r (4 + 2r) / 2 = r (2 + r)
Geometric Progression
Quantities are said to be in geometric progression (G.P.) when they increase or
decrease by a constant factor. The constant factor is called the common ratio and is
calculated by dividing any term by the term that immediately precedes it.
For example,
9, 18, 36, 72...
½, -1, 2, -4, 8...
In the above example, the first series is multiplying by 2, a common ratio of +2 and the
second series is increasing and decreasing alternatively by -2, i.e., a common ratio of
-2.
The general form of a geometric progression is
a, ar, ar2, ar3 , ar4 ..........
where ‘a’ is the first term of the series and ‘r’ the common ratio (ar2/ar = r)
It can be noticed from the general series that the index of r is one less than the number
of the term in series:
For example, 2nd term = ar
3rd term = ar2
10th term = ar9
23
Therefore the nth term in a G.P. is given by

Tn = arn-1
As the common ratio is a constant factor obtained by dividing any term by its
preceding term, it can be expressed in the general form as
r = tn / tn-1 where tn is the nth term of the series.
To find the sum of number of terms in geometrical progression
Consider the general form of a Geometrical Progression i.e.
a, ar, ar2,.........., arn-1
Here ‘a’ is the first term of the series, ‘r’ the common ratio and arn-1 is the nth term of
the series.
Let ‘S’ denote the sum of the G.P.
∴ S = a + ar + ar2 +...+ arn-1 → (1)
Multiplying every term by r
⇒ rS = ar + ar2 + ar3 +...+ arn → (2)
Subtracting (1) from (2)
⇒ rS- S = ar+ar2+ar3+.....+arn - a-ar-ar2-ar3....-arn-1
= arn - a
⇒ (r-1)S = a (rn-1)
⇒ S = a (rn-1) / ( r-1) → (3)
Changing the signs of the numerator & denominator
⇒ S = a (1-rn) / (1-r)
(a − ar n )
or S =
(1 − r )
Since the nth term of a G.P. is tn = arn-1
⇒ S = (rtn-a) / (r-1) → from (3)
a (rn - 1) / (r-1) = (arn-a) / (r-1)
= r (tn – a/r) / (r-1)
= (tn – a/r) / (1-1/r)
Example 2.31
The sum of 5 terms of a geometric progression is 3905 and the common ratio is 5. Find
the first term of the series.
The sum of geometric series with n terms is given by
a (r n − 1)
S=
(r − 1)
a (55 − 1)
3905 =
(5 − 1)
15620
a=
3125 − 1
∴a = 5
24
Example 2.32
The third term of a geometric progression is 9 and the seventh term is 729. Find the
sum of the first 10 numbers.
Solution:
The third term of the geometric progression is given by ar2 and the seventh term is
given by ar6 .
ar 2 9
=
ar 6 729
1 1
=
r 4 81
∴r = 3
ar2 = 9
9
∴a = =1
32
The sum of terms in the geometric progression is given by
a (r n − 1)
Sn =
(r − 1)
1(310 − 1)
=
(3 − 1)
= 29524
Geometric Mean
Suppose three quantities are in G.P, then the middle one is said to be the geometric
mean (G.M.) between the other two.
∴ If a, ar, ar2 are in G.P. Then ‘ar’ is the G.M.
Suppose a, G, b are in G.P. Then
G/a = b/G (Common ratios are equal)
⇒ G2 = ab ⇒ G = ab
∴ The Geometric mean, G = ab
If ‘n’ geometric means are inserted between ‘a’ and ‘b’ such that the whole series then
formed will be in G.P, then there are( n+2) terms in the series including the first term
‘a’ and last term ‘b’.
⇒ b is the (n + 2)th term
⇒ b = arn+1
⇒ r = (b/a)1/(n+1)
Therefore the required means between a and b are
ar, ar2 ...arn where r = (b/a)1/(n+1)
Kth G.M. = a (b/a)K/(n+1)
Product of n GMs = (ab)n/2
25
Example 2.33
Insert four geometric means between 1/5 and 1 / 15625.
Solution:
Given that there are six terms in the series and the 1st term is 1/5 and last term is
1/15625. Let r denote the common ratio.
∴ 6th term = ar5
⇒ 1/15625 = ar5
⇒ 1/15625 = 1/5 (r)5
⇒ r5 = 1/3125
⇒ r = 1/5
∴ The means are 1/25, 1/125, 1/625, 1/3125.
If ‘r’<1, then greater the value of ‘n’, smaller will be the value of rn and hence arn/(1-r)
tends to 0. Therefore for a series where n tends to infinity, the value arn/1-r is
considered to be negligible and the sum of the series is given by
a
Sn =
(1 − r )
Harmonic Progression
Quantities are said to be in harmonic progression (abbreviated H.P.) if the series
obtained by taking reciprocals of the corresponding terms of the given series is an
arithmetic progression.
For example, the series 1, 1/4, 1/7, 1/10,... is an H.P. since the series obtained by taking
reciprocals of its corresponding terms i.e. 1, 4, 7, 10, ..... is an A.P.
A general H.P. is 1 / a , 1 / ( a + d ) , 1 ( a + 2d ) ,......
nth term of an H.P. = 1 / [ a + ( n –1 ) d ]
Example 2.34
The 7th term of an H.P. is 1/10 and 12th term is 1/25. Find the 20th term, and the nth
term.
Solution
Let the H.P. be 1 / a , 1 / ( a + d ) , 1 ( a + 2d ) , ...
The 7th term = 1 / (a + 6d) = 1 / 10 => a + 6 d = 10
The 12th term = 1 / (a + 11 d) = 1 / 25 => a + 11 d = 25
Solving these two equations, a = -8, d = 3
Hence 20th term = 1 / ( a + 19d ) = 1 / [ -8 + 19 ( 3 ) ] = 1 / 49
and nth term = 1 / [ a + ( n - 1) d ] = 1 / [ -8 + ( n – 1 ) 3] = 1 / [ 3n – 11 ]
Example 2.35
Prove that three quantities a, b, c are in A.P., G.P., or H.P. if
(a-b) / (b-c) = 1, a / b or a / c respectively.
Solution
a, b, c are in A.P. if b -a = c -b i.e.
if (a-b) / (b-c) = 1 = a/a
Also a, b, c are in G.P.
26
If b/a = c/b i.e if 1 - b/a = 1 - c/b

i.e. if (a-b) / a = (b-c) / b
i.e. if (a-b) / (b-c) = a/b
Similarly a, b, c are in H.P.
if 1/b - 1/a = 1/c - 1/b
i.e. if (a-b) / ab = (b-c) / bc
i.e. if (a-b) / (b-c) = ab / bc = a/c
Harmonic mean
Three numbers a, b and c are in H.P., if 1/a, 1/b, 1/c are in A.P. It means,
1/b - 1/a = 1/c - 1/b
i.e. if 1/a + 1/c = 2/b
i.e. if b = 2ac / (a + c)
Thus the H.M. between a and c is H = 2ac / (a + c)
The harmonic mean of the positive real numbers a1,...,an is
n
H=
1 1 1 1
+ + ............. +
a1 a2 a3 an
The harmonic mean is never larger than the geometric mean or the arithmetic mean. If
A.M., G.M. and H.M. are the arithmetic mean, geometric mean and harmonic mean of
a set of numbers respectively,
A.M. ≥ G.M. ≥ H.M.
In certain situations, the harmonic mean provides the correct notion of "average". For
instance, if for half the distance of a trip you travel at 40 miles per hour and for the
other half of the distance you travel at 60 miles per hour (i.e. less time), then your
average speed for the trip is given by the harmonic mean of 40 and 60, which is 48. it
is discussed in detail in Chapter 5.
Example 2.36
The Harmonic mean of a series 1/8, 1/12 and 1/16 is
Solution:
n
Harmonic mean is given by H =
1 1 1 1
+ + ............. +
a1 a 2 a 3 an
3 48 144
= = 3× =
1 1 1 13 13
+ +
8 12 16
FUNCTIONS
Many questions in management are studied by constructing mathematical models of
the underlying phenomenon. We may regard a model as a collection of relationships
among economic variables. Functions are a convenient way of stating such
relationships. The purpose of this section is to introduce some elementary
mathematical concepts and techniques that are employed in business analysis.
27
For a layman, “function” indicates a relationship among objects. A function provides a

model to describe a system. Managers refer to demand functions, which indicate the
sales volume of an item, as a function of its price. Similarly, economists refer to
supply function which considers production volume of an item as a function of the
prevailing/projected price of the item.
A function expresses the relationship of one variable or a group of variables (called the
domain) with another variable (called the range) by associating every member in the
domain to a unique member in range.
Suppose X represents the “price of a good” and Y the “demand”. We may postulate
that Y is related to X in the sense that if we fix the price of the good, then we will be
able to determine the demand. We say that Y is a function of X since we are able to
compute a unique value of Y for a given value of X. We may represent the
relationship as Y = f(X), where f represents the relationship. It is important to note that
it may be the case, though it is not necessary, that the relationship is a causal one, that
is, X is the cause and Y is the effect. When the relationship is causal, we may regard X
as the independent variable and Y as the dependant variable.
Thus, y = f(x) = 4x-5 and y = g(x) = 3x2-2x+100 are examples of functions. But y2 = x
is not a function of x since the rule that a given value of x should yield a unique value
of y is violated. The following example will make the concept of functions more clear.
Suppose we play a game involving the toss of two fair coins. And for every head (H)
that turns up, you win Re.1 and for every tail (T) that turns up, you lose Re.1
Let D = {TT, HT, HH} and R = {-2, 0, 2}.
Then the game may be represented by the function
R = f (D),
where f (TT) = -2, f (HT) = 0 and f (HH) = 2.
Let us consider another example.
Suppose a variable Y is related to a variable X as
Y = 5 X +3
For any given value of X, the value of Y can be calculated using the above
relationship. Thus, the equation enables us to construct a range of Y values for a given
table of X values.
Functions of Many Variables

There may be more than one independent variable which determines the value of the
dependent variable y. The dimension of a function is determined by the number of
independent variables in the domain of a function. For example:
y = f(x, z) is a two dimension function.
y = f(x1, x2,...xn) is an n-dimensional function.
Linear Functions
Linear functions are of the form:
y = a0+a1x1+a2x2+........+anxn
where a0, a1, a2... an are constants and x1, x2 ...xn are variables.
In two dimensional space, a linear function is a straight line and is usually written as y
= a + mx.
The Y-intercept is ‘a’, and the slope of the line is ‘m’.
28
If m > 0: As x increases, y increases. The line rises from left to right. This is illustrated
in the Figure 2.1
If m < 0: As x increases, y decreases. The line falls from left to right.
If m = 0: y = a, is a line parallel to the X-axis.
If m = ]: x = a, a line parallel to y-axis.
Figure 2.1: Linear Functions
y y
m negative
m positive
x x
negative slope
positive slope
y
y
m undefined
m=0
x x
undefined slope
zero slope
Equation of a straight line

In the two dimensional case, the form of the linear function can be obtained if we know
the co-ordinates of two points on the straight line. Suppose x’ and x” are two given x
values and the corresponding y values are y’ and y”, then the slope of the line is:
y '− y ' '
m=
x '− x ' '
The intercept is
y '− y ' '
a = y '−[ ]x'
x '− x ' '
In the ‘n’ dimensional case, we need the co-ordinates of at least ‘n’ points for finding
the linear functional form.
Example 2.37
Find the line relating Sales (Y) to Advertising expenditure (X) if
X: 20 23 26 29
Y: 160 172 184 196
29
Solution:
If we assume that the 4 points (20,160), (23,172), (26,184) and (29,196) will fall on a
straight line, then we may compute ‘a’ and ‘m’ using any pair of points.
m = (172-160)/ (23-20) = 4, and
a = 160 - (4 × 20) = 80, and
y = 80 + 4x
One may verify that all 4 points fall on the line. For example, consider the point
(29,196).
80 + (4×29) = 196.
Inverse Functions
We have seen that if y is a function of x, then for each given value of x, we can
determine uniquely the value of y as per the functional relationship. For some
functions, it is possible to express x in terms of y so that, given the value of y, the value
of x can be uniquely determined. The function that expresses the variable x in terms of
the variable y is called the inverse function and is denoted by x = f-1(y).
If y = f (x) = 4x + 5
Solving x in terms of y
y−5
x= = f −1 ( y )
4
Exponential Functions
The exponential functions are useful for describing topics like compound interest and
growth. The exponential function is defined as:
y = m.ax
where ‘m’ and ‘a’ are constants with ‘a’ (called the base) which takes only a positive
value and ‘x’ is an independent variable.
The exponential curve rises to the right for a > 1 and m > 0 and rises to left for a < 1
and m > 0. If x takes only positive integral values (1, 2, 3...), y = max is the xth term in a
Geometric Progression.
Logarithmic Functions
If x = by where both b > 0 and x > 0, then we define y = logbx, which is read as “y is
the log of x to the base b”. So a log is simply the inverse function of an exponential
function. Therefore, the domain of any logarithmic function is the set of positive
numbers since the range of any exponential function is the set of positive numbers.
The exponential function and logarithmic function are inverse functions of each other,
since
y = logbx if and only if x = by.
If y = log10x = f(x), then x = 10y and vice-versa.
If x = ey then y = lnx and vice-versa.
SIMULTANEOUS EQUATIONS
Many models are formulated as a set of equations involving a number of variables. The
simplest of these are simultaneous linear equations. The idea is to find one or more set
of values for the variables which will satisfy the system of equations.
30
Example
Find the equilibrium point if the supply and demand functions are:
S(x) = 2x + 1 and D (x) = 14/(x + 2).
Here x denotes the price of the good, S(x) describes the quantity that the producers are
willing to supply at the given price and D(x) describes the consumer demand at the
given price. The equilibrium point is the price at which supply equals demand.
That is, 2x + 1 = 14/ (x +2)
or, 2x2 + 5x – 12 = 0
− b ± b 2 − 4ac
x=
2a
By using the above formula we get,
− 5 ± 5 2 − 4(2)(−12)
=x =
2(2)
− 5 ± 25 + 96
=x =
4
x=6/4 or – 16/4.
Since the price cannot be negative, x = 1.5. One may verify that this value of x satisfies
both the demand and the supply functions.
System of Linear Equations

As mentioned above, the simplest of such problems is a system of linear equations.
This system can be solved mechanically by the diagonalization method. This method is
used in the simplex method which one will come across in a branch of mathematics
called linear programming.
The method creates a series of 1s across the diagonal, and 0s elsewhere. The method is
explained through the examples below.
Example
Solve
2x – y – z = 2
x+y +z=1
x + y + 2z = 1
We may represent the above system of equations by the table below. For example, the
last row of the table reads: lx + ly + 2z = 1, which is exactly the last equation given
above.
The first row and the first column are merely headings. We will therefore number the
rows and columns from 0 onwards. Therefore the third equation is represented by the
third row, columns 1 to 4.
31
TABLE 0
EQUATION X Y Z RHS
1 2 -1 -1 2
2 1 1 1 1
3 1 1 2 1
The idea is to progressively introduce 1 in the diagonal cells (Row 1 and Column 1,
Row 2 and Column 2, Row 3 and Column 3) and 0s elsewhere using only the
following operations
• Multiply a row by any constant
• Add a multiple of any row to any other row
We introduce 1 in Cell (1,1) and 0s in the remaining cells of column 1 by
Multiplying Row 1 by 0.5 and then subtracting the modified Row 1 from Rows 2 and
3. That is,
Row 1: Row 1 * 0.5
Row 2: Row 2 + (-1 * Row 1)
Row 3: Row 3 + (-1 * Row 1)
TABLE 1
EQUATION X Y Z RHS
1 1.0 -0.5 -0.5 1.0
2 0.0 1.5 1.5 0.0
3 0.0 1.5 2.5 0.0
Now we want to introduce 1 in Cell (2,2) and 0s in the other cells of Column 2. To
achieve this, we need to multiply Row 2 by (1/1.5). Follow this by adding 0.5 times
Row 2 to Row 1, and adding -1.5 times Row 2 to Row 3.
TABLE 2
EQUATION X Y Z RHS
1 1.0 0.0 0.0 1.0
2 0.0 1.0 1.0 0.0
3 0.0 0.0 1.0 0.0
We now introduce 1 in Cell (3, 3) and 0s elsewhere in Column 3. But note that Cell (3,
3) is 1 and Cell (1, 3) is 0 already. To achieve our objective, we need Cell (2, 3) = 0.
This is done by adding -1 * Row 3 to Row 2.
TABLE 3
EQUATION X Y Z RHS
1 1.0 0.0 0.0 1.0
2 0.0 1.0 0.0 0.0
3 0.0 0.0 1.0 0.0
Since we cannot go any further in the diagonalization process, let us construct the
equations implied by Table 3.
32
1x + 0y + 0z = 1.0
0x + 1y + 0z = 0.0
0x + 0y + 1z = 0.0
One solution is immediate: x = 1, y = 0 and z = 0
NOTE
A system of solutions may have
• Exactly one solution
• More than one solution
Our method will generate one of the solutions or imply that there is no solution.
Example
Solve
x+y=2
-2x – 2y = 3
TABLE 0
ROW X Y RHS
1 1 1 2
2 -2 -2 3
We must make Cell (2, 1) = 0. This is done by adding (2*Row 1) to Row 2.

TABLE 1
ROW X Y RHS
1 1 1 2
2 0 0 7
We may immediately stop since Row 2 of Table 1 reads
0x + 0y = 7. That is,
0 = 7!
And this is impossible. Therefore the system of equations has no solutions.
Example
Acme Limited is planning to manufacture two toy trucks: Mini and Jumbo. It takes two
hours to produce a batch of Minis and three hours to produce a batch of Jumbos. Each
batch of Minis takes 3 tons of material 1 and each batch of Jumbos take 7 tons of
material 1. Each batch of Minis takes 3 ton of material 2 and each batch of Jumbos
takes 2 tons of material 2. Suppose there are 7 hours in a day and for each day, 13 tons
of material 1 and 8 tons of material 2 are available. How many batches of each toy
should be produced to utilize exactly all available labour and material resources?
Let x represent the number of batches of Minis that should be produced, and y the
number of batches of Jumbos.
2x + 3y = 07 (Labor)
3x + 7y = 13 (Material 1)
3x + 2y = 08 (Material 2)
33
TABLE 1
EQUATION X Y RHS
1 2.0 3.0 7.0
2 3.0 7.0 13.0
3 3.0 2.0 8.0
TABLE 2
EQUATION X Y RHS
1 1.0 1.5 3.5
2 0.0 2.5 2.5
3 0.0 -2.5 -2.5
TABLE 3
EQUATION X Y RHS
1 1.0 0.0 2.0
2 0.0 1.0 1.0
3 0.0 0.0 0.0
The solution can now be read off table 3.
x = 2 and y = 1
That is, produce 2 batches of Minis and 1 batch of Jumbos.
SUMMARY
This chapter intends to provide the student a fundamental knowledge of mathematics.

It discusses the basics of mathematics starting from the types of number systems used,
concepts of permutations and combinations, logarithms, progressions, and functions.
The two number systems that we generally use are decimal and binary systems. This
chapter also discussed the octal number system.
Permutations refer to the arrangements that can be made by taking some or all of a
number of objects available. Combinations refer to selections that can be made by
taking some or all of a number of things available. This chapter discussed the concepts
of permutations and combinations.
The logarithm of any number to a given base is the index or the power to which the
base has to be raised in order to equal the given number. This chapter discussed the
concepts of logarithms.
Progression is a series of numbers which follow a specific rule. According to the
relationship between the constituent numbers, progressions are of three types-
Arithmetic, Geometric and Harmonic. This chapter also discussed the concepts of
progressions and functions. The different types of functions are linear functions,
inverse functions, exponential functions and logarithmic functions.
34
Chapter 3
Calculus
• Limits
• Derivatives
• Optimization
• Integral Calculus
LIMITS
The concept of a limit is fundamental in calculus. Often, we are interested in
knowing the behavior of f(x) as the independent variable x approaches some
particular point ‘a’. The question arises, that if we give values to x which are nearer
and nearer to ‘a’, will the values of f(x) come nearer and nearer to any particular
value? Suppose we define a function f(x) as:
f(x) = 2x
It can be seen that as we give values to x which are nearer and nearer to 0, then the
value of f(x) also comes nearer and nearer to 0.
If as x approaches a value ‘a’, f(x) approaches some number L, then we say that the
limit of f(x) as x approaches a is L. This is symbolically written as
lim f(x) = L
x→a
x → a is to be read as ‘x approaches a’ or ‘x tends to a’
Sometimes we may allow x to take values which are larger and larger, without any
limit. This is symbolically written as x → ∞ (read as ‘x approaches infinity’). If f(x)
approaches a limit L as x → ∞ , then we write
lim f(x) = L
x→∞
In some cases, it may so happen that as x approaches a value, then the value of the
function f(x) may become larger and larger without any limit. This is symbolically
written as:
lim f (x ) = ∞
x→L
Example
Suppose f(x) = 2x2 – 1
As x approaches value 1, f(x) approaches the value 1,
lim f(x) = 1
x →1
This is graphically represented below.
1
x
-1
36
Calculus
Example
Let f(x) = (x+1)/x.
From the table and graph given below, we see that f(x) approaches 1 x → ∞ as but
never equals 1.
x y = f(x) = (x+1)/x
1 2
10 1.1
100 1.01
1000 1.001
10,00,000 1.000001
1
x
x +1
Thus lim =1
x
x→∞
Example
1
Suppose f(x) = , what will happen to f(x) as x approaches 1? The table and
(x − 1)2
graph for this function are given below.
y
x y=f(x)=1/(x-1)2
0 1
1/2 4
2/3 9
4/5 25
9/10 100
999/1000 10,00,000
x
1
The value of f(x) becomes larger and larger, as x approaches 1.

1
Thus lim →∞
(x − 1)2
x→∞
37
DERIVATIVES
The rate of change in the value of a function is useful in studying the behavior of a
function. The change in y for a unit change in x is referred to as the derivative of y
with respect to x. In finance and economics, the rate of change is termed marginal or
incremental. For example, the marginal cost of capital is the rate of change of the
total cost of capital per unit change in new capital raised.
The idea of the derivative as the rate of change of the function at a fixed point has a
geometrical foundation. The slope of the tangent to the function at the point equals
the derivative at the point.
d df
The derivative is usually denoted by of f(x) or or f′(x) It may be noted that
dx dx
the derivative itself is a function, and the value of the derivative depends on where it
is evaluated.
The derivative of a function f(x) at point ‘a’ is defined as:
f ( x ) − f (a )
lim
x−a
x→a
The process of getting the derivatives is called ‘differentiating’ a function. There are
some rules of differentiation which are given below:
Rules of Differentiation
a. If f(x) = c, where c = a constant, then f ′( x ) = 0 . Slope of a horizontal line is
zero, since by definition the value of y does not change when the value of x
changes.
n −1
b. If f ( x ) = x n then f ′( x ) = n. x
c. Let f(x) be a product of two functions, say, g(x) and h(x) so that
f(x)=g(x).h(x), then
f ′( x ) = g ′( x ).h ( x ) + g ( x )h ′( x )
d. If f(x) is a quotient of two functions, say, g(x) and h(x) so that
g(x)
f(x) = , where h(x) ≠ 0
h(x)
g ′(x)h(x) − g(x)h ′(x)
Then f ′(x ) =
[
h (x )
2
]
e. If f(x) = c.g (x) where c is a constant, then,
f ′(x ) = c.g ′(x )
f. If f (x ) = g (x ) + h (x ), then
f ′(x ) = g ′( x ) + h ′(x )
⎛1⎞
g. If f (x ) = In x, then f ′(x ) = ⎜ ⎟
⎝x⎠
g (x ) g (x )
h. If f(x) =e , then f ′(x) = g ′(x).e
g ′(x )
i. If f(x) =In g(x), then f ′(x ) =
g(x )
38
Calculus
Examples
f ( x ) = 4x 4 − 2x 2 + 7 x + 5
a f ′(x ) = 4
d x4 ( )
− 2.
d x2
+ 7.
( )
d(x ) d(5)
+
dx dx dx dx
( )
= 4 4 x − 2(2x ) + 7(1) + 0
3
= 16 x 3 − 4 x + 7
b (
f(x) = x 3 − 4 x 2 + 1 )( )
f ′(x) =
(
d x3 − 4 2 )(
x + 1 + x3 − 4 )(
d x2 +1
)( )
dx dx
= 3x 2 (x 2 + 1) + x 3 − 4 ( 2
d x +1
)( )
dx
= 3x (x + 1) + x − 4 (2x )
2 2 3 ( )
= 5x 4 + 3x 2 − 8x
f ( x ) = x −3
c
3
f ′( x ) = −3x − 4 = −
x4
2x 3 + x + 3
d f (x ) =
x2 +1
(x 2
+1) dxd (2x 3
) (
+ x + 3 − 2x 3 + x + 3 ) dxd (x 2
)
+1
f ′(x ) =
( x 2 + 1) 2
( x 2 + 1)(6 x 2 + 1) − (2x 3 + x + 3)(2x )

=
( x 2 + 1) 2
2 x 4 + 5x 2 − 6 x + 1
=
2
⎛⎜ x 2 + 1⎞⎟
⎝ ⎠
2
f(x) = e (3 x + x )
e
d (3x 2 + x )
= (3x 2 + x ) e
dx
2
= (6 x + 1) e(3x + x )
Function of a Function
Suppose y is a function of z,
y = f (z)
and z is a function of x
z = g (x).
39
Since, y depends on z, and z in turn depends on x, y is also a function of x.

Thus y = f (z)
= f [g(x)]
The derivative of y with respect to x can be obtained as:
dy dy dz
=
dx dz dx
= f ' ( z )g ' ( x )
The above method is often used for getting the derivatives of some complicated
functions.
Example
2 5
Let y = (2x + 1)
The derivative of y with respect to x could be obtained by expanding the right hand
side expression and taking the derivative term by term. This is a long and laborious
process. The derivative could be easily obtained by using the concept of function of a
function.
Let us define a function z as follows:
2
z = 2x +1
then, the function y is y = z5
dy
= 5.z 4
dz
dz
Also = 4x
dx
dy dy dz
Hence, =
dx dz dx
= (5.z 4 ).(4x)
= 5.(2x 2 + 1) 4 . (4x)
= 20x (2x 2 + 1) 4
Example
2x + 1
y =
x+2
2x + 1
Let y = μ where μ =
x+2
d
dμ
(x + 2) (2x + 1) − (2x + 1) d (x + 2)
= dx dx
dx (x + 2) 2
=
(x + 2)(2) − (2x + 1)(1)
(x + 2)2
40
Calculus
2x + 4 − 2x − 1 3
= =
(x + 2) 2
(x + 2)2
1
− 1
dy / dμ = 1 / 2.μ 2 = 1 / 2.
μ
3 1 1
dy / dx =
2
(x + 2) 2 μ
3 1 x+2
dy /dx =
2 2 2x + 1
(x + 2)
3 x+2 1
=
2 2 x + 1 ( x + 2) 2
Higher-Order Derivatives
It can be seen that the derivative of a function is also a function.
Considering f′(x) as a function of x, we can take the derivative of f′(x). which will
yield another function; say f”(x). This is called the second derivative of f (x). The
third derivative of f(x) is the derivative of the second derivative f′(x). We can take
derivatives of higher orders by repeating the process. The second, third, fourth, etc.
derivatives are symbolically written as;
2
d f
f ″ (x) or
2
dx
3
d f
f ″′ (x) or
3
dx
4
(4) d f
f (x) or
4
dx
n
(n) d f
f (x) or n
dx
Example
Let f(x) = 3x 3 + 2x 2 − 4x + 6
f ′(x) = 9x 2 + 4x − 4
f ″(x) = 18x + 4
f ″ (x) = 18
(4)
f (x) =0
(5)
f (x) =0
...........................
(n)
f (x) =0
41
Example
Let y = x 3 − 9x 2 + 9x
What is the value of d 2 y/dx 2 for x = 1?
y = x 3 − 9x 2 + 9x
dy/dx = 3x 2 − 18x + 9
d 2 y / dx 2 = 6x − 18
d 2 y/dx 2 (at x = 1) = 6 ×1 − 18 = −12
Remark
Of all the higher order derivatives, the second order derivative is of special interest
especially in optimization.
The first derivative of a function, f′(x), is the slope of the function f(x), or the rate of
change in the value of f(x) per unit change in x. Similarly the second derivative, f″(x)
is the slope of the function f′(x), or the rate of change in the value of f′(x) per unit
change in the value of x, i.e., it is the rate of change of the original function f(x).
Definition
A function f(x) is said to be monotonically increasing if its first derivative, f′(x) is
positive for all values of the independent variable x. It is said to be monotonically
decreasing if f′(x) is negative for all values of x.
Examples
a. Consider the linear demand function
p = 100-10q, where
p = price per unit sold (average revenue),
q = quantity sold
The total revenue function, T (q), is obtained as p*q.
Thus T(q)=(100-10q)q=100q-10q2
The first derivative gives the relationship of the change in total revenue per unit
change in quantity sold. This is called the marginal revenue, MR.
dT(q)
MR = = 100 − 20q
dq
The changes in marginal revenue for changes in quantity sold can be obtained from
the slope of the marginal revenue curve. This is the second derivative of T(q).
d
(MR ) = d 2T(q ) = −20
dq dq 2
The marginal revenue declines at a constant rate of 20, per unit increase in quantity
sold. The relationships are shown below.
b. In developing the theory of investor choice under uncertainty, cardinal utility
functions, U(R) are used. These utility functions should have the property,
U′(R)>0 and U″ (R) < 0. Check this property for the following utility functions:
42
Calculus
T (q) MR or P
Total Revenue = 100 MR = 100 – 20q

100q – 10q2 P = 100 – 10q
75
250 50
150 25
0 5 10
0 5 10 q
i. U (R) = aR – bR2 where a and b are constants

ii. U (R) – In R
aR
iii. U (R) = - e – where a is a constant
OPTIMIZATION
Optimization is an activity which aims at finding the best (optional) solution to a
problem.
Situations frequently arise in finance and other areas where optimization is required.
Organizations would like to maximize their profits or minimize their costs at a given
level of output. An individual would like to maximize his utility when choosing
among investment alternatives. If we have a mathematical function, then we can find
a solution to the optimization problem using calculus.
Here, we have to make a distinction between relative maxima (or minima) and global
maxima (or minima). let f (x) be a function of x. Then the global maxima of f(x) is
the maximum value which f(x) could take for all the possible range of value of x. For
example, if x denotes the number of units produced and f(x) denotes the total profit
when x units are produced, then we are interested in the maximum value of f(x) for x
in the range 0 to the maximum production capacity. Compared to the global
maximum, we speak of relative maxima, which is the maximum value of f(x) in a
relatively small range of values of x. f(x) is said to have a relative maxima at a point
f (x)
f (c)
f (a)
+ + d
a - x
-
f (b)
f (d)
43
x = a, if the value of the function at a, f(a) is greater than or equal to f(x) for values of
x sufficiently close to value a. Look at the graph of a function f(x) given below.
This function has a relative maxima at point x = a and at x = c, and a relative minima
at points x = b and x = d. If we consider values of x in the range 0 to d, then global
maxima of the function is f(c) attained at point x = c.
The following characteristic features of the relative maxima and minima could be
observed:
i. For values of x which are sufficiently close to a (or c), the value of the function
first increases and then decreases. The slope of the function is positive up to a
(or c) and then it becomes negative. For points b and d, the slope changes from
negative to positive.
ii. The tangent lines at points a, b, c and d are horizontal, indicating that the slopes
at those points are zero.
From the above observations, the following results could be obtained:
Result 1: if f(x) has a relative maxima or minima at x = a, then the value of the
derivative of f(x) at a, f′(a) is 0.
Note that the converse may not be true. If at a point a, f′(a) = 0, it does not
imply that f (x) has a relative maxima or minima.
Result 2: If, at point a, we have f′(a) = 0 and f″ (a) <0 then f(x) has a relative
maxima at x = a.
If f′(a) = 0 and f″ (a) > 0 then f(x) has a relative minima at x = a.
This is because at relative maxima (or minima) the slope changes from
positive (or negative) to negative (or positive) for points x close to a.
If f′(a) = 0 and f″(a) = 0 then we cannot say anything definitely about the
occurrence of maxima or minima at point a.
Example
A manufacturer fixed the following demand curve:
P(x) = 200-3x
Where x = quantity and p(x) = price at that quantity.
Let the fixed cost of production be Rs. 25 and the variable cost Rs. 2 per unit. The
manufacturer wants to fix the level of output so as to maximize his profit.
The profit function f(x) is:
f(x) = total revenue – total cost
= (200-3x) x – (25 +2x)
= -3x2 +198x – 25
The first order derivative is
f′(x) = - 6x +198
198
This is 0 if – 6x + 198 = 0 or x = = 33
6
At this level of output, i.e. x = 33, we would like to examine whether we have a
minimum or maximum. The second order derivative is:
f″(x) = -6 which is negative for all values of x
Therefore f″(a) < 0.
44
Calculus
Hence, we conclude that at the level of production of 33 units, the manufacturer will
make a maximum profit.
The maximum profit is:
2
- 3 × (33) + 198 × 33 − 25 = Rs. 3,242.
Suppose, the quantity that could be produced is limited to the range, say, 10 to 30
units, then we have to check the value of the profit at the boundary points 10 and 30,
in addition to any other points in the range (10,30) and then find the optimal level of
production.
Partial Derivatives
So far we have considered differentiation of functions of one independent variable. In
many situations, we come across functions with more than one independent variable.
Since the value of the function is influenced by each independent variable, the rate of
change in the value of the function relative to the change in one independent variable
can be studied by holding the other independent variables constant. Let z = f(x, y).
The change in z for changes in x can be obtained by holding y constant. This is the
basic idea behind partial differentiation. The rules for partial differentiation and
ordinary differentiation are exactly the same except that when the partial derivative
of one independent variable is taken, the other independent variables are treated as
constant. The partial derivatives of a function f(x, y) are symbolically represented by
∂f ∂f
, to indicate the partial derivative with respect to x and the partial derivative
∂x ∂y
with respect to y.
Example
Let w(x, y, z) = xy + yz + zx
∂w
then, =y + 0+ z = y + z
∂x
∂w
= x + z + 0= x+ z
∂y
∂w
=0+ y + x= y+ x
∂z
NOTE
If a function has two independent variables x and y so that
Z = f(x, y)
Then, the condition for relative maximum or minimum is:
∂f ∂f
= 0, = 0,
∂x ∂y
2
ii ∂ 2f ∂ 2f ⎡ ∂ 2f ⎤
<⎢ ⎥ for maximum
∂x 2 ∂y 2 ⎢⎣ ∂x∂y ⎦⎥
2
∂ 2f ∂ 2f ⎡ ∂ 2f ⎤
>⎢ ⎥ for minimum
∂x 2 ∂y 2 ⎢⎣ ∂x∂y ⎥⎦
45
INTEGRAL CALCULUS
This branch of mathematics concerns itself with an operator called the integral
∫ f(x) dx is called the indefinite integral of f(x)
b
∫ f (x)dx is called the definite integral of f(x).
a
Indefinite Integral
We may regard the indefinite integral as an inverse operator of differentiation.
d
Therefore ∫ f(x)dx = F(x) if F(x) = f(x)
dx
Note that
d d
F(x) = f(x) implies [F(x) + C] = f(x) for any constant C.
dx dx
Example
2
x
F(x) =
1 2
2
x
F′(x) = x ⇒ ∫ xdx =
2
2x
2 F(x) = e
2x 2x 2x
F′(x) = 2e ⇒ ∫ 2e dx = e
F(x) = In (x)
1 1
3 F′(x) = ⇒∫ dx = In x
x x
4 x2
F(x) = + 10
2
x2
F′(x) = x ⇒ ∫ xdx = + 10
2
5 F(x) = e 2x + 3
F′(x) = 2e 2x ⇒ ∫ 2e 2x dx = e 2x + 3
6 F(x) = In (3x) = In 3 + In x
1 1
F′(x) = ⇒∫ dx = In x + In 3
x x
∫ f(x) dx = F(x) + C
where, C is any constant,
and,
d
F(x) = f(x)
dx
46
Chapter 4
Arranging Data
• Meaning of Data
• Types of Data
• Data Collection
• Tables and Graphs as Data Presentation Devices
• Frequency Distribution
In organizations managerial decisions are based on vast primary and secondary data.
The raw data available to the managers is mostly in an unorganized form. The first task
of the manager in decision making process is to get the data (figures) properly
organized, classified, and presented in a meaningful manner in order to facilitate
decision making. Presenting raw data in a meaningful way requires two steps to be
carried out: classification of data and tabulation of data. In this chapter we will discuss
the objectives of classification and various devices of data presentation like tables,
graphs, frequency distribution, histograms, frequency polygons, and cumulative
frequency curves.
MEANING OF DATA
Data is a collection of related observations, facts or figures. Managers collect facts

such as number of units produced per day by each department or by each worker, and
these facts and figures of production can be called as data. A collection of data is called
a data set, and each observation a data point.
TYPES OF DATA
Managers rely on data for decision making. This data can be broadly grouped into two
categories:
• Published data (data that is already collected and published) and,
• Unpublished data (data that is yet to be collected or printed).
Published data can be obtained in the form of bulletins, reports etc. from various
Government agencies. For example, data relating to monetary and banking activities
(industrial production index, price index, etc.) can be obtained from Reserve Bank
bulletins and finance reports published by Reserve Bank of India every month. Apart
from government agencies, many research organizations and private sources provide
useful data to managers.
In some cases, published data may not be available to managers to make decisions. In
such cases, managers have to go in for first hand data collection by the way of sample
survey or a census. The information may be obtained by using various methods like
observation, personal interview or questionnaires. The data can also be grouped as
primary or secondary data. The first hand data that is collected by the researcher
personally is primary data and that collected from other available sources (collected by
others) is known as secondary data. Large volumes of data are available within every
organization in the form of financial statements, sales reports, cash flow data,
production schedules, budgets, etc. Managers can make decisions based on this data in
the light of data relating to the industry and economy as a whole.
Examples of Raw Data

Information before its systematic arrangement and analysis is called raw data. For
example the number of units produced for each production run in a month, number of
bags of paddy produced per acre of land owned by an Agricultural Research Institute
etc.
Table 4.1: Output for 30 Production runs (in 000 tons)
24.5 20.8 23.0 24.5 23.8 23.9 22.8 21.6 24.0 26.4
23.6 22.8 23.4 25.1 22.8 24.3 22.2 18.3 24.8 24.6
24.6 24.5 23.6 24.0 24.2 23.8 24.8 22.9 23.0 23.8
48
Arranging Data
Table 4.1 shows the output levels for 30 production runs. The information a manager
gets from the table is an example of raw data. Managers can derive useful inferences
from this raw data by applying various statistical methods. For instance managers can
use this above raw data to find a pattern in production output and also the average out
put per production run.
DATA COLLECTION
Managers select the samples in such a way that all related cross sections of the
population are represented in the data. Population is the entire collection of entities that
a manager is trying to study and sample is a fraction of the population that represents
the entire population in its characteristics proportionately.
For example, when a magazine conducts an opinion poll among 5000 individuals from
all over India, with a view to know the general opinion of Indians towards politicians,
then all Indians is the target population and 5000 individuals represent the population
and it is referred to as a sample.
To estimate the potential market for a new innovation, for example, managers in
research department may study 500 consumers in a particular territory. The managers
make it sure that this sample contains the consumers belonging to all cross sections
(income, religion, education, and locality) of the society.
Managers collect relevant data from either first hand observations or from already
existing records. For example, a welfare manager gets the details about the families of
the employees either from the records or from a fresh survey. This information is then
arranged to produce data that enables easy decision making. When data is arranged in a
crisp, usable form; managers can take reliable information from the available internal
data and the external environment to make intelligent decisions.
Before using the data, managers have to test the data for its adequacy and reliability, as
it influences the quality of the final decision. Managers can pose the following
questions to test the validity of the data:
1. Where does the data originate from?
2. Is the source reliable?
3. Does the data support or contradict the previous decisions?
4. Are the conclusions derived from the data?
5. What is the size of the sample? Does it represent the entire population under
consideration for decision making?
When the manager gets the answers to all the above questions, he can easily improve
the quality of his decisions. The quality of the data collected depends on the sample
selected from the entire population.
Methods of Collecting Data

Though data can be collected using various methods, they can be broadly classified
into two types:
• Complete enumeration and
• Sample method
Complete enumeration is a method in which the entire population is taken up and
information is collected relating to all the units of the population. This is known as
Census Survey or Census. Governments undertake complete enumeration of all the
citizens once in every ten years and the information is collected in respect of each and
49
every person living in India. In the same way organizations undertake census of the
employees’ families to collect the information that helps designing of welfare schemes.
Sample method is a method in which enumeration of a part of the population or
universe is taken up and information is gathered regarding this selected part. This is
commonly adopted in organizations to check the quality of the finished products of a
manufacturing unit.
Managers choose one of the above data collection methods depending on the factors
like purpose of the enquiry, time available for making a decision, budget allocation,
and the accuracy of data required for decision making.
Classification of Data
Classification is the process of arranging data into sequences and groups according to
their common characteristics. For example data collected in a consumer survey can be
classified along characteristics like age, gender, education, income, etc.
Some common methods of classification are
• Geographical, i.e. area-wise or region-wise
• Chronological,
• Qualitative, i.e., depending on characteristics
• By magnitude.
For example, the data relating to profits of a XYZ company for the past 10 years can be
classified chronologically as shown in Table 4.2.
Table 4.2: Net Profits of XYZ Company (in $ millions)
Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Profit 191 154 90 -10 80 200 246 310 360 390
TABLES AND GRAPHS AS DATA PRESENTATION DEVICES
Data collected by various methods can be presented in the form of tables and graphs. In
tables the data is classified based on time of observation, magnitude or some other
characteristics of the variable. On the other hand graphs are pictorial representation of
the data.
Tabular Presentation of Data

Tabulation is the logical listing of related quantitative data in vertical columns and
horizontal rows with sufficient explanatory and qualifying words, phrases and
statements in the form of titles, headings and explanatory notes to make clear the full
meaning, context and origin of the data.
Tabular presentation is used to summarize or condense data. Tables help the managers
to analyze the relationships and trends in the collected data.
Objectives of tabulation
Some of the objectives of tabulation are
• To condense complex data
• To show a trend
• To display huge volumes of data in less space
• To be self explanatory
50
Arranging Data
• To highlight key characteristics of data

• To facilitate comparison of data elements
• To help decision making using statistical methods
• To serve as reference for future decisions.
Parts of an ideal table
The purpose of a good table is to represent all the available data in the least possible
space without losing its clarity. These tables must be self explanatory to the maximum
extent possible. The main parts of a table are as follows:
1. Table Number - Acts as an identity to the table
2. Title - Gives an idea about the nature of data in the table
3. Captions - These are headings given to vertical column(s) that explains the mode
of classification. i.e. time, quantity, region etc.
4. Stubs - These are the headings explaining the basis for classifying the rows.
5. Body - The data posted in rows and columns, where row and column headings
(captions & stubs) explain the data.
6. Footnote - Any other information to explain the data in the table.
7. Source - Source of information.
Table 4.3 gives a better illustration of these parts.
Table 4.3: Sales data of Alpha Electronics, for the Month of May and June
2000 (in lakhs Rs.)
Territory (Caption)
Product East Zone West Zone North Zone South Zone

(Stub) (Sub-Cap 1) (Sub-Cap 2) (Sub-Cap 3 ) (Sub-Cap 4)
May June May June May June May June
T.V. 11 14 8 10 20 16 16 14
Audio 8 6 14 20 12 8 10 12
Graphical Presentation of Data

In graphical presentation, the collected data is represented by various types of
geometrical devices such as points, lines, bars, multi-dimensional figures, pictorials,
etc. A graphical method is a non-quantitative form of presentation; the quantities are
also indicated along with them. The magnitude of the data is depicted visually through
the proportional size of the diagram or graph. Thus, graphs, maps, charts are drawn to
“scale” to give accurate data to managers.
Several types of graphs or charts are used to present data. Some of them are:
1. Line charts
2. Bar charts
3. Pie charts
4. Pictograms
5. Scatter diagrams
51
Line charts
Line chart is one of the effective graphical methods to depict the trend in a data. For
example, a manager can easily notice the trend in the sales data when presented in a
graph than in a tabular form. Figure 4.1 shows the line chart for the data depicted in
Table 4.4.
Table 4.4: Sales of Mars Corp. Ltd. 1990-1999 (in lakhs Rs.)
Year Sales
1990 300
1991 360
1992 330
1993 390
1994 400
1995 390
1996 420
1997 460
1998 460
1999 480
Figure 4.1: Line Chart
550
500
450
400
350
Sales
300
250
200
150
100
50
0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Year
Bar charts
Business and statistical data can be presented through bar charts. Bar charts use
rectangles to present the data. These diagrams are one-dimensional because the
magnitude of the data is represented by length of the bar. Bar charts consist of a group
of equidistant rectangular bars, each representing one class interval of a given data in
the table. The thickness or width of the bar has no relevance, only the length of the bar
52
Arranging Data
is important. Bar charts are the easiest and most flexible general purpose charts used
for data representation.
A bar chart can be drawn in different forms: vertical or horizontal. In horizontal charts
the bars originate from the Y-axis but for vertical bar charts they originate from the X-
axis
Guidelines for constructing a bar chart are as follows:
1. Length of the bars should be proportionate to the data they represent.
2. All bars should rise from the same base line.
3. Uniformity in the width of the bars need not be maintained as width is irrelevant in
bar charts.
4. The scale should be selected depending on the highest value in the table.
5. The gaps between the bars should be uniform.
6. The bars should be arranged from left to right.
7. It is desirable to mention the (magnitude) figure on the bar to enable the manager
to get the magnitude at a glance.
Now consider the data relating to sales of Mars Corp. is given in Table 4.4. When we
represent the sales values of last five years (1995 to 99) in a bar chart it appears as in
Figure 4.2.
Figure 4.2: Bar Chart for Table 4.4 (1995-1999)
600
460 460 480
420
Sales
400 390
200
0
1995 1996 1997 1998 1999
Year
Multiple bar diagram or compound bar diagrams are used to compare two or more sets
of related data. This diagram is similar to the simple bar diagram, but bars in each set
are placed together and gap is left between each set of bars. Table 4.5 shows the export
and import value of Bharat Petroleum Corporation Limited (BPCL) for 4 consecutive
years between 1995 and 1998. Figure 4.3 is compound bar chart showing the export,
import values of BPCL.
Table 4.5: Export Import Values of BPCL
1995 1996 1997 1998
Exports 2720 3610 4170 3570
Imports 430 870 710 1780
53
Figure 4.3: Compound Bar Chart Showing the Export and Import
Values of BPCL
5000
Export/Import
4000
3000 Export
2000 Import
1000
0
1995 1996 1997 1998
Year
Pie diagrams
Pie diagram is a circle divided into various segments and each segment represents the
percentage contribution of various components to the total. For drawing a pie diagram
it is necessary to express the value of each category as a percentage of the total. 3600 in
a circle represent the whole (i.e., 100%) and 3.60 constitute 1% of the total. Managers
use pie diagrams to compare many components simultaneously.
Table 4.6 depicts the data on share holding pattern in BPCL. Figure 4.4 presents the
same in the form of a pie diagram.
Table 4.6: Shareholding Pattern of BPCL

Category of share holders Holding (in %)
Indian promoters 66
Indian institutions / Mutual funds 20
FIIS 10
Public 4
Total 100
Degree of each part = Part × 360/Total = Part × 3.6
Figure 4.4: PIE Diagram Representing Share Holding Pattern in BPCL
PIE CHART
4% Indian Promoters
10%
IND'I /MF's
MFs
20% FIIS
FIIs
Public
66%
Pictogram
Pictograms represent the data in the form of pictures. The data is presented using
appropriate pictures and their sizes indicate the magnitude of the data.
54
Arranging Data
Scatter diagram
Scatter diagram is used to study the correlation between two dependent variables.
Production manager can study the relation between the cost and the level of production
to find the optimal production level. Figure 4.5 shows the scatter diagram showing
relationship between cost and volume. The cost of production is plotted on the graph at
various levels of production, the obtained points are scattered and hence it is known as
scattered diagram. When the points on the graph follow a pattern, it indicates high
correlation and irregular pattern or behavior indicates low correlation.
Figure 4.5: Scatter Diagram

Y
4.00
3.75
3.50
3.25
Cost
3.00
2.75
2.50
2.25
2.00 X
50 55 60 65 70 75 80 85 90 95
Volume of Production
FREQUENCY DISTRIBUTION
The table in which raw data is tabulated by dividing it into classes of convenient size
and computing the number of data elements (or their fraction out of the total) falling
within each pair of class boundary is called a frequency distribution table. Frequency
distribution has two parts, the left of the table has sizes or magnitude of values and on
its right the number of times a value or a group of values are repeated.
The raw data relating to the ages of 50 employees of a department are as given in Table
4.7
Table 4.7: Ages of 50 Employees
56 30 36 28 43 52 25 29 38 40
48 57 33 29 32 48 46 49 32 31
39 28 52 21 47 28 39 44 54 30
22 33 48 39 56 27 54 28 36 32
40 46 26 35 45 55 28 31 43 46
As the ages in the table are in a arbitrary manner it is very difficult for the human
resources manager to grasp any trend from the raw data. Logical arrangement of data is
necessary to compress the data and help the manager know to which age groups an
employee belongs and what is the frequency of each age group.
55
Construction of Frequency Distribution

To construct a frequency distribution, the data is to be divided into groups of similar
intervals. Then the number of data points that fall into each group has to be recorded
against each group. This frequency distribution gives the valuable information at a
glance regarding the pattern in the ages of the employees (i.e. information like which
age group most of the employees belong to? How many employees belong to each age
group, etc.) Once the data is presented in the form of a frequency distribution, it shows
the trend or pattern of data and it no longer shows the exact value of each data element.
A frequency distribution is a tabular form that organizes data into classes i.e. groups of
values having same characteristics of data. Thus, a class refers to a group of objects
with a common property. The range of values of a given class is called a class limits,
and middle of a class interval is called class mark.
Table 4.8 shows the frequency table for the data depicted in Table 4.7. In this
frequency distribution table the data is divided into eight classes (the number of classes
can vary depending on the need for comparison). The class interval in this frequency
distribution is five (24-20=5, 29-25=5.....etc) and values 20, 24 are the limits of the
first class 25, 29 are limits of second class and so on. The frequency distribution shows
the number of observations from the given data that fall into each of the above classes.
Thus when the frequency with which each value occurs in each class is known, a
frequency distribution can be constructed.
Table 4.8: Frequency Distribution Table Presenting Ages of 50 Employees
Class Frequency
20-24 2
25-29 10
30-34 9
35-39 7
40-44 5
45-49 9
50-54 4
55-59 4
Frequency distributions can be constructed with classes of qualitative attributes. For
example Table 4.9 shows the educational qualifications of 100 employees in the form
of a frequency table.
Thus classification can be either quantitative or qualitative and either discrete or
continuous classes.
Histograms
A histogram is a series of rectangles, the width of each being proportional to the range
of values within a class and height being proportional to the number of items falling in
the class. The widths of the bars are uniform when the widths of classes in a frequency
distribution are equal. And the length of the bar is proportionate to the number of data
elements in the class it represents.
56
Arranging Data
Table 4.9: Educational Qualification of 100 Employees

Educational Frequency Relative frequency
Qualification distribution distribution
Matriculates 36 0.36
Plus two 14 0.14
Science graduates 20 0.20
Arts graduates 12 0.12
B. Tech 14 0.14
Post graduates, PhD 4 0.04
Total 100 1.00
Figure 4.6: Histogram for Data Shown in Table 4.8

12
10
Frequency
4
2 10 9 7 5 9 4 4
2
20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59

Age
When a histogram is constructed using relative frequency, it is called a relative
frequency histogram. While the absolute histogram represents the number of data
items, the relative frequency histogram shows the relative size of each class with the
total. A histogram for the data in the frequency distribution Table 4.8 is given in Figure
4.6.
FREQUENCY POLYGONS
Frequency polygons represent graphically both simple and relative frequency
distributions. For constructing a frequency polygon, the frequencies are marked on the
vertical axis and the values of variables (that are being studied) are taken on the
horizontal axis (as done for a histogram). Dots are put on the graph against the class
marks to represent the frequencies. These dots are connected by drawing straight lines,
this forms a frequency polygon. When the straight line are smoothed by adding classes
and data points, is called a frequency curve. The number of classes and the
observations are increased and the dots are connected by curves to get a frequency
curve.
Cumulative Frequency Distribution
A cumulative frequency distribution is a tabular display of data showing how many
observations lie above, or below, certain values, rather than simply showing the
number of items within intervals (as shown in frequency distribution Table 4.8. For
example, if the manager wants to know the number of employees of age less than 40
years he can use the cumulative (less-than) frequency table shown in Table 4.10.
57
Table 4.10: Cumulative (less than) Frequency Distribution

Table Presenting Ages of 50 Employees
Class (Years) Cumulative Frequency
Less than 25 2
Less than 30 12
Less than 35 21
Less than 40 28
Less than 45 33
Less than 50 42
Less than 55 46
Less than 60 50
When the cumulative frequencies are plotted on a graph we get an Ogive. A less-than
Ogive of the distribution of ages of 50 employees is shown in Figure 4.7.
The points on the graph represent the number of employees having less age than the
number of years shown on the X-axis.
When an Ogive is drawn for a cumulative (more than) frequency distribution the graph
would slope down and to the right, instead of up as shown in Figure 4.7.
Figure 4.7: “Less than” Ogive of the Distribution of 50 Employees
60
50
40
Cumulative
30
Frequency
20
10
0
Age < 30 35 40 45 50 55 60
25
Skewness and Kurtosis

Skewness and Kurtosis are the two characteristics of data sets that provide useful
trends and patterns in the data represented as frequency distribution curves.
Skewness is the extent to which a distribution of data points is concentrated at one end
or the other; or the lack of symmetry in the curve. The curves representing the data
points in the data set can be of two types:
• Symmetrical curves
• Skewed curves (positively or negatively skewed)
58
Arranging Data
A curve is said to be symmetrical when a vertical line drawn from the center of the
curve to the X-axis divides the area under the curve into equal parts. A symmetric
curve is shown in Figure 4.8.
Figure 4.8: Symmetric Curve
A curve is said to be skewed when the values in the frequency distribution are
concentrated more towards the left or right side of the curve i.e. the values are not
equally distributed from the center of the curve. A curve is said to be positively skewed
when the tail of the curve is more stretched towards the right side. It is said to be
negatively skewed when the tail is more stretched towards the left side. The skewed
curves are shown in Figure 4.9.
Figure 4.9: Skewed Curves

Negatively
Positively skewed
skewed
Kurtosis is the degree of peakness of a distribution of points i.e. Kurtosis measures the
peakedness of a distribution. Two curves with same central location and dispersion
may have different degrees of kurtosis (See Figure 4.10).
Figure 4.10: Curves with Different Kurtosis but same Central

Location
Curve A Curve B:
59
SUMMARY
Data is a collection of related observations, facts or figures. This data can be broadly
grouped into two categories: Published data and unpublished data. Managers use
sampling methods to draw inferences about the data. Population is the entire collection
of entities that a manager is trying to study and sample is a fraction of the population
that represents the entire population in its characteristics proportionately. Data
collected by various methods can be presented in the form of tables and graphs. In
tables, the data is classified based on time of observation, magnitude or some other
characteristics of the variable. On the other hand graphs are pictorial representation of
the data. Several types of graphs or charts are used to present data. Some of them are
line charts, bar charts, pictograms, pie charts and scatter diagrams. This chapter also
discussed how frequency tables can be used to represent data.
60
Chapter 5
Measures of Central
Tendency
• Objectives of Averaging
• Requisites of a Good Average
• Types of Averages
• Mathematical Averages
• The Median
• The Mode
The most important objective of a statistical analysis is to calculate a single value that
represents the characteristics of the entire available raw data. This single value
representing the entire data is called the ‘central value’ or an ‘average’. This value is
the point around which all the other values of the data cluster. Therefore it is known as
measure of location and since this value is located at a central point nearest to other
values of the data it is also known as measure of central tendency. This chapter
discusses various measures of central tendency like mean, median and mode and their
use in day to day management activities. For example, the mean sales of a territory
give a rough idea to the sales manager about the performance of the sales personnel in
that territory.
OBJECTIVES OF AVERAGING
The major objective of averaging is to arrive at a single value that is representative of

the entire data. Some of the objectives of averaging are as follows.
1. To find out one value that represents the whole mass of data
The objective of averaging is to represent a set of individual values in a concise way,
so that the researcher can have an instant idea about the size of each entity in the group.
Averages help the researcher or manager to grasp the characteristics of the data group
without studying every value in the group. For example, a manager need not look at
ages of every trainee of a fresh batch if the average age is calculated by dividing the
total age of all the trainees by number of trainees. This average is a value that enables
the manager to have a overall idea about the characteristics of the huge volumes of
data.
2. To enable comparison
Averages act as a common denominator for comparing two or more sets of data. They
also help in drawing conclusions about the characteristics of different sets of data. For
example, a manager can use the average sale of two territories to compare the
performance of sales executives of two territories. These average sales figures of each
territory reduce the burden of going through the volumes of sales data to know the
performance of each territory. Thus, a quick and easy comparison of sales of the two
territories is made possible for a manager by these averages.
3. To establish relationship
Averages play a major role in establishing relationships between separate groups in
quantitative terms. It is vague if one states that productivity of an employee in Wipro is
more than that of average productivity of Satyam Computer Solutions employee. It
would be clearer if both the productivities are expressed in terms of averages.
4. To derive inferences about a universe from a sample
Averages help a manager to get valuable inferences about the whole universe by means
of sample data. The average calculated from a sample data gives a reliable idea about
the average of the entire universe.
5. To aid decision-making
Averages act as benchmarks or standards for managerial control and decision-making.
A production manager may rely on average employee productivity to set future
production targets for individuals and the organization as a whole. Thus these averages
(average turnover, etc.) act as benchmarks for performance appraisal and decision-
making in future.
62
Measures of Central Tendency
REQUISITES OF A GOOD AVERAGE
An ideal average should have the following characteristics:

• Should be rigidly defined
• Should be mathematically expressed (Have a mathematical formula)
• Should be readily comprehensible and easy to calculate
• Should be calculated based on all the observations
• Should be least affected by extreme fluctuations in sampling data.
• Should be suitable for further mathematical treatment.
In addition to the above requisites, a good average should also retain maximum
characteristics of the data, it should be a nearest value to all the data elements.
Averages should be calculated for homogeneous data i.e. ages, sales etc.
TYPES OF AVERAGES
Averages or measures of central tendency are of the following types:

1. Mathematical averages
• Arithmetic mean
• Geometric mean
• Harmonic mean
2. Positional averages
• Median
• Mode
Of the above, arithmetic mean, median and mode are the widely used averages. Figure
5.1 shows the types of averages.
Figure 5.1: Types of Averages
Averages
Mathematical averages Positional averages
Geometric Harmonic
Arithmetic Median Mode
mean mean
mean (A.M.) (Md) (Mo)
(G.M.) (H.M.)
63
MATHEMATICAL AVERAGES
Arithmetic Mean
The arithmetic mean or mean is the most simple and frequently used average.
Arithmetic mean is represented by notation x (read x - bar).
Calculating the Mean from Ungrouped Data

The mean x of a collection of observations x1 ,x2,................., xn is given by:
x = (1/n) (x1 +x2 ...+ xn)
= ∑x/n
n
= (1/n) ∑ x i
i =1
where, x is sample mean

i is the set of natural numbers
n
∑ x is the sum of values of all observations
i =1
n is the number of elements.

∑ indicates that all the values of x are summed together.
When the mean is calculated for the entire population it is known as population
arithmetic mean (µ). ‘N’ is the number of elements (observations) in the population.
Then µ = ∑ x/N
Example 5.1
Table 5.1: Absentee List of Drivers of the Transport Department over a
Span of 90 Days
Driver 1 2 3 4 5 6 7 8 9 10
Number of days on leave 8 6 6 7 4 5 6 2 4 7
Solution:
When a manager wants to know the average number of days a driver is on leave in 90
days, he can calculate the mean of the ungrouped data as follows:
x = ∑ x/n = 8 + 6 + 6 + 7 + 4 + 5 + 6 + 2 + 4 + 7
10
= 55/10
= 5.5 days per driver out of 90 days
In the above example, the mean is calculated by adding every observation separately,
in no set order. This is an ungrouped data. One can calculate the mean using the above
method for limited values. But the task becomes difficult while calculating average for
a vast data, say for 5000 employees. In such cases a frequency distribution of the data
will be helpful to a manager, and mean should be calculated using a different method.
Calculating the Mean from Grouped Data (Frequency distribution)

A frequency distribution consists of data that are grouped into classes. Every
observation (value) is placed in one of the classes. Unlike the earlier example (Table
5.1) the manager is unaware of the individual values of every observation of the
64
universe. Say a Finance manager wants to find out the average monthly pay of 600
employees in an organization, and he is having a frequency distribution (shown in
Table 5.2). He can easily calculate an estimate of the value of the mean of this grouped
data. It is referred to as estimate as he need not use all the 600 observations. The
estimate of the value of mean will not be as accurate as the value obtained by
computing from all the 600 observations (i.e. by adding all the values)
Table 5.2: Average Monthly Pay of 600 Employees

Class (Rupees) Frequency
1000 - 2999 50
3000 - 4999 110
5000 - 6999 162
7000 - 8999 100
9000 - 10999 83
11000 - 12999 45
13000 - 14999 25
15000 - 16999 15
17000 - 18999 8
19000 - 20999 2
Total 600
To compute the arithmetic mean of grouped data, calculate the midpoint of each class
and multiply each mid point (class mark) by frequency of observations in the
corresponding class. Then he has to add all these results, and divide the sum by the
total number of observations.
Mid point (class mark) = x = (lower limit + upper limit)/2
The formula for computing Arithmetic mean for grouped data is:
x = ∑ (f × x)/n
Where, ∑ = Notation for “Sum”
f = Number of observations in each class
x = class mark (mid point of each class)
n = Number of observations in the sample
Table 5.3 shows the steps in calculating arithmetic mean for grouped data.
Table 5.3: Calculation of Arithmetic Mean for Grouped Data

Class (Rs) Class Mark (X)* Frequency (f) (f)× (X)
(1) (2) (3) (2) × (3)
1000-2999 2000 50 1,00,000
3000-4999 4000 110 4,40,000
5000-6999 6000 162 9,72,000
7000-8999 8000 100 8,00,000
9000-10999 10000 83 8,30,000
11000-12999 12000 45 5,40,000
13000-14999 14000 25 3,50,000
15000-16999 16000 15 2,40,000
17000-18999 18000 8 1,44,000
19000-20999 20000 2 40,000
65
n = 600, ∑ (f × x ) = 44, 56, 000
Sample mean, x = ∑ (f × x )/n = 44,56,000/600 = Rs 7426.66

* Class mark adjusted to nearest integers
In the above example, the approximate mean (average salary) is Rs. 7426.66.
Let us compare the actual mean and the appropriate mean calculated for data in Table
5.1.
x = ∑ x/n
= 8+6+ 6+ 7+ 4+5+ 6+ 2+ 4+ 7
10
= 55/10 = 5.5 days
Now let us calculate the appropriate mean for the same data presented in a frequency
distribution as follows:
Table 5.4: Calculation of Arithmetic Mean for Ungrouped Data

Class (Rs) Class Mark Frequency f×X
(1) (X) (2) (f) (3) (3) × (2)
0- 2 1 1 1
3- 5 4 3 12
6- 8 7 6 42
∑ (f × x ) = 55
x = ∑ (f × x )/n
= 55/10 = 5.5 days
The values of actual mean calculated from ungrouped data and appropriate mean
calculated from grouped data are equal in case of above example Table 5.1. But these
two values may not be equal all the time. There can be minor difference between the
means when the number of observations increases.
Short-cut method of computing arithmetic average for a grouped data (using
codes):
The Shortcut method for computing mean is very useful to managers when the number
of classes is more and the classes are of equal size.
Steps for calculating mean using shortcut method:
1. Locate an assumed mean A. Though any point in the grouped data can be selected
as mean A, for simplicity, it is selected from one of the centrally located classes.
Assign a ‘code value’ zero to the selected class.
2. Assign negative integers as codes to the values smaller than the class mark and
positive integers to larger values as follows:
Class 1-10 11-20 21-30 31-40 41-50 51-60 61-70
Code -3 -2 -1 0 1 2 3
↑
x
66
3. Assign ‘x’ to represent the class mark that is assigned the code “zero”. The
following formula is used to determine the sample mean using codes:
x = x0 + w ∑ (u × f)/n
Where,
x = mean
x0 = value of the class mark assigned the code 0
w = numerical width of the class interval
u = code assigned to each class
f = frequency of the class (number of observations)
n = total number of observations in the sample.
Table 5.5 shows how to code the class marks and find the sample mean for the average
monthly income of employees.
Table 5.5: Average Monthly Income of an Employee
Class
Class (Rs) Code (u) Frequency (f) u×f
Marks (X)
(1) (3) (4) (3) × (4)
(2)
1000 – 2999 2000 -4 50 -200
3000 – 4999 4000 -3 110 -330
5000 – 6999 6000 -2 162 -324
7000 – 8999 8000 -1 100 -100
9000 – 10999 10000 0 83 0
11000 – 12999 12000 1 45 45
13000 – 14999 14000 2 25 50
15000 – 16999 16000 3 15 45
17000 – 18999 18000 4 8 32
19000 – 20999 20000 5 2 10
n = 600 ∑f = n = 600 ∑ u×f = -772
x = x0 + w ∑ (u × f)/n
=10,000 + 2,000 x (-772)/600
= 10,000 – 2573.33
= Rs.7426.66 ≈ 7427 (average monthly salary)
Advantages and Disadvantages of Mean

The first advantage of arithmetic mean is that its concept is familiar and clear to most
people. The second advantage is that it is easy to understand and easy to calculate.
Every data set has one and only one mean. Finally, arithmetic average provides a good
basis for comparison. For example, if a manager wants to compare the performance of
salesmen of four different regions of a state, arithmetic average provides the correct
basis for assessing the relative efficiency of the regions.
However, Arithmetic mean suffers from the following draw backs. First, although the
mean is representing all the values in the data set, it may be affected by the highly
67
fluctuating values that are not far from other values of the group. Observe that if the
units produced in a day by 5 workers of a batch as in Table 5.6 the mean units
produced per day is
Table 5.6: Number of Units Produced by Workers in a Day
Worker 1 2 3 4 5
Units 23 22 24 21 5
µ = ∑ x/n = (23 +22 +24+21+5) / 5 = 19 units
When the mean units are calculated leaving the fifth worker (i.e. 5) the mean is 22.5
units. Thus, one extreme value ‘5’ has affected the mean. Hence, it is more appropriate
to calculate the mean excluding the extreme value in order to make it more
representative.
The second disadvantage is that it is very difficult to find the actual mean (using µ =
∑x/n). Third disadvantage is that we cannot calculate the mean for a data set with
open-ended classes at either end of the scale. A class that allows either the upper or
lower end of a quantitative classification scheme to be limitless is called as open-ended
class.
The Weighted Arithmetic Mean

The weighted mean is calculated taking into account the relative importance of each of
the values to the total value. Consider, for example, the manufacturing company in
Table 5.7 that employs three grades of labor (unskilled, semiskilled, and skilled) to
produce two products. When the company wants to know the average cost of labor per
hour for each product, the simple arithmetic average of the labor wage will be:
Table 5.7: Labor - Capital Involved in Manufacturing Two Products
Wage per hour (x) Labour hours per unit
Class of Labour
(Rs) Product 1 Product 2
Unskilled 10 2 6
Semiskilled 15 3 2
Skilled 20 5 1
x =∑x/n
= (10 + 15 + 20)/3 = Rs. 15 / hour
When the above average wage per hour is taken to calculate the labor cost of one unit
of Product 1, the value would be
15 (2 + 3 + 5) = Rs. 150
And, labor cost for one unit of Product 2 is
15(6 + 2 + 1) = Rs. 135
But these values calculated using simple arithmetic average are incorrect as it does not
take into consideration the fact that different amounts of each class of labor are used.
The correct value (cost per one unit) can be determined in the following manner.
For Product 1, the total labor cost per unit
= (10×2) + (15×3) + (20×5)
= Rs. 165
and, cost per hour
= Rs. 165/(2 + 3 + 5)
=Rs. 16.5/hour
68
For product 2, the total labor cost per unit

= (10×6) + (15×2) + (20×1)
= Rs. 110
and, cost per hour
= Rs. 110/ (6 + 2 + 1)
= Rs. 12.22 per hour
An alternative method to calculate the correct average cost per hour for the products is
to take a ‘weighted average’ of the cost of the three classes of labor. For doing this, the
hourly wage, for each class of labor is given a “weight” in proportion of total labor
required to produce the product.
One unit of Product 1 required 10 hours of labor
Unskilled labor = 2/10 of 10 hours
Semi-skilled labor = 3/10 of 10 hours
Skilled labor = 5/10 of 10 hours
When these above fractions are used as weights, then Cost of labor (per hour) for
product 1 is:
= (2/10 × 10) + (3/10 × 15) + (5/10 × 20)
= Rs. 16.5 / hour
Similarly, for Product 2 cost of labor (per hour) for 1 unit is:
= (6/9 × 10) + (2/9 × 15) + (1/9 × 20)
= Rs. 12.22 / hour
∴ The formula for calculating the weighted average is:
x w = ∑(w × x)/ Sw
where,
x w = symbol for weighted mean
w = weight allocated to each observation (2/10, 3/10, 5/10 for product 1 in
the above example)
∑ (w×x) = sum of each weight multiplied by that element.
Sw = sum of all the weights
Now, apply this formula to find the labor cost / hour for Product 1 for the above
problem
xw = ∑ (w×x) / Sw
= {(2/10 × 10) + (3/10 × 15) + (5/10 × 20)} / {2/10 + 3/10 + 5/10}
= Rs 16.5/1
= Rs. 16.5/ hour
Example 5.2
Rohan Prasad bought shares worth Rs. 2400 every month for five months. Per share
price of shares for the first, second, third, fourth and fifth month was Rs. 10, Rs. 12, Rs
15, Rs. 20 and Rs. 24 respectively. What is the average price that Rohan has paid at the
end of fifth month?
69
The number of shares bought in the first month = 2400/10 = 240
Similarly for the second, third, fourth and fifth month is 200, 160 120 and 100
respectively.
240 200 160 120 100

Therefore weigthed average = × 10 + × 12 + × 15 + × 20 + × 24
820 820 820 820 820
= 14.63
Example 5.3
The following information is available for two data
x1 = 10 x 2 = 12
n1 = 6 n 2 = 10
Calculate the combined mean of both the samples.
Solution:
N1µ1 + N 2µ 2
µ=
N1 + N 2
(6 × 10) + (10 × 12)

=
6 + 10
= 11.25
Example 5.4
The average monthly income of employees of Krishika Manufacturing Co. was Rs
2250. The total number of employees was 75. While auditing, it was found that two
entries were recorded incorrectly. Rather than Rs 2150 and Rs 2500, they were
recorded as Rs 2200 and Rs 2000. Which of the following represents the correct
monthly average of employees?
ΣX = 2250 × 75 = 168750
Sum of incorrect values = 2200 + 2000 = 4200
Sum of correct values = 2150 + 2500 = 4650
Correct ΣX = 168750 – 4200 + 4650 = 169200
Therefore, correct average = 169200/75 = 2256
Example 5.5
The average mark of both girls and boys in the class of business strategy is 59. If the
average marks of boys are 65 and that of girls are 50. The percentage of boys and girls
respectively in the class is
N 1 X1 + N 2 X 2
X12 =
N1 + N 2
X12 = 59, X1 = 65 and X2 = 50, N1 + N2 = 100
N × 65 + (100 - N1 ) × 50
∴ 59 = 1
100
⇒ 15 N1 + 5000 = 5900
⇒ N1 = 60
1N2 = 100 – 60 = 40
70
Geometric Mean
Managers often come across quantities that change over a period of time, and may
need to know the average rate of change over a period of time. Arithmetic mean is
inaccurate in tracing such a change. Hence a new measure of central tendency is
needed to calculate the change rate-the “Geometric Mean”.
G.M = n product of all the values
= n x1 × x 2 × .......x n = (x1 × x2 × x3 ... xn)1/n
where ‘n’ is the number of values.

Geometric mean is applicable in many cases. Its use in calculating the growth rates of a
textile unit in the southern region for the last five year are given below:
Table 5.8: Growth Rate of Textile Units

Year 1 2 3 4 5
Growth rate (%) 7 8 10 12 18
The geometric mean
= n x1 × x 2 × .......x n
Where,
x1, x2,… xn are termed as the growth factor and is equal to 1+ (rate/100)
GM = 5
1 . 07 × 1 . 08 × 1 . 10 × 1 . 12 × 1 . 18 = 1.1093
1.1093 is the average growth factor. The growth rate is calculated as
1.1093 – 1 = 0.1093
Then, 0.1093 x 100 = 10.93
So the growth rate is 10.93 percent per year.
Harmonic Mean
Harmonic mean is based on the reciprocals of numbers averaged. It is defined as the
reciprocal of the arithmetic mean of the reciprocal of the given individual observations.
Thus, by definition:
HM = N
1 1 1
+ + .... +
X1 X 2 Xn
Where, X1, X2, X3, etc. refer to various items of the variable and N refers to the total
number of items.
Example 5.6
Gopi walks from his house to the bank at the speed of 2 kmph, while returning from
the bank to the house, his speed was 3 kmph. Calculate the average speed for the whole
walk.
71
N
Average speed =
1 1
+
x1 x 2
2 12
= = = 2.4
1 1 5
+
2 3
Example 5.7
Matel Plastics Ltd. got a raw material delivery order from Blowplast Inc. However, the
condition was that the delivery had to be made within four hours, failing which the
order would be considered cancelled. Robert, the salesman at Matel, was assigned the
responsibility to make the delivery. Robert had to be careful not to exceed the 80 kmph
speed limit, otherwise he would be flouting the traffic rules. The marketing manager
asked him not to go below 60 kmph as there was a risk of the order being cancelled.
Robert divided his journey into four equal parts. He traveled the first quarter of the
distance at the speed of 50 kmph, the second quarter of the distance at 65 kmph and the
last quarter of the distance he covered at the speed at 55kmph.
He was successful in delivering the product on time. If his average speed was 60.5
kmph, what was his speed when he covered the third quarter of the distance?
Solution:
Assume that Robert covers the first, second, third and fourth quarter of the distance at
the speed of X1, X2, X3 and X4 respectively.
Let the average speed of Robert’s whole journey from Matel to Blowplast be Ho =
60.5.
From the given information in the problem, we have
X1 = 50
X2 = 65
X4 =55
Ho = 60.5 and N = 4.
After inserting the values in the formula for calculating the harmonic mean we get:
H= N
⎡ 1 1 1 ⎤
⎢ + + .... + ⎥
⎢⎣ X 1 X2 Xn ⎥⎦
60.5 = 4
⎡ 1 1 1 1 ⎤
⎢ + + + ⎥
⎣⎢ 50 65 X 3 55 ⎥⎦
⇒ 1 ⎡ 1
⎢ +
1
+
1
+
1 ⎤
⎥ = 1 ⎡⎢ 0.0535664 + 1 ⎤⎥ = ⎡ 1 ⎤
4 ⎢ X ⎥ ⎢ 60 . 5 ⎥
4 ⎢⎣ 50 65 X3 55 ⎥⎦ ⎣ 3⎦ ⎣ ⎦
⎡ 1 ⎤
⇒ 1 [0 .0535664 ] + 1
⎢ ⎥ = 0.0165289
4 4 ⎣⎢ X 3 ⎥⎦
⇒ 0.0133916 + 1 = 0.0165289
4X 3
72
⇒ 1 = 0.0031373
4X 3
⇒ 1 = 0.0125492
X3
⇒ X3 = 79.68 ≈ 80
Thus, in the third hour Robert traveled at the speed of 80 kmph.
THE MEDIAN
The median, as the name suggests, is the middle value of a series arranged in any order
of magnitude.
As it is distinct from the arithmetic mean which is calculated from the value of every
item in the series, it is called a positional average. The term ‘position’ refers to the
place of a value in the series. The median is just the 50th percentile value below which
50% of the values in the sample fall. The object of median is therefore not merely to
fix a value that shall be representative of a set, but also to establish a dividing line
separating the higher from the lower values.
Calculating the Median from Ungrouped Data

If the data set contains an odd number of items, the middle item of the array is the
median. If there is an even number of items, the median is the average of the two
middle items. If the total of the frequencies is odd, say n, the value of (n+1)/2th item
gives the median and when the total of the frequencies is even, say, 2n, then nth and (n
+ 1)th are two central items and the arithmetic mean of these two items gives the
median.
Example 5.8
The following data relates to the sales figures of certain companies relating to the year
1992-93:
Table 5.9: Sales Figures of Companies
Companies Sales
ACC 1520
Andhra Valley 436
Excel Inds 228
Indian Hotels 239
Tata Hydro 292
Tata Power 734
Tata Tea 412
Voltas 980
Tomco 312
Tinplate Co. 256
73
Solution:
The median for the above data can be obtained as follows:
The series should first be arranged in an order. In the present case, it has been arranged
in descending order.
Table 5.10: Sales Figures of Companies Arranged in Descending Order
Company Sales Rank
ACC 1520 1
Voltas 980 2
Tata Power 734 3
Andhra Valley 436 4
Tata Tea 412 5
Tomco 312 6
Tata Hydro 292 7
Tinplate Co. 256 8
Indian Hotels 239 9
Excel Inds 228 10
As there are 10 elements, the median will be the mean of the 5th and the 6th items, i.e.
(412 + 312)/2= 362
Thus, the median sales value of the ten companies is 362.
Calculating the Median from Grouped Data

In order to find the median, first the median class is to be located and then interpolation
is to be used by assuming that items are evenly spaced over the entire class interval.
The formula used for the calculation of median is
⎡ ((N + 1)/2) − (F + 1) ⎤
Median = ⎢ ⎥ W + Lm
⎣ f m ⎦
where,
Lm = lower limit of the median class
fm = frequency of the median class
F = cumulative frequency up to the lower limit of the median class
W = width of the class interval
N = total frequency.
Example 5.9
Let us find median for the following data.
Table 5.11: Gross Profit as a Percentage of Sales
Gross Profit as a
0-10 10-20 20-30 30-40 40-50
percentage of sales
No. of companies 21 32 43 34 23
74
Table 5.12: Cumulative Frequency

Gross Profit (%) No. of Companies (f) Cumulative Frequency (cf)
0-10 21 21
10-20 32 53
20-30 43 96
30-40 34 130
40-50 23 153
Here the total frequency N = 153.

(N + 1) (153 + 1)
Median is the size of the th item, i.e. th item, i.e. the size of the 77th
2 2
item. It lies in the class 20-30. Hence 20-30 is the median class, of which the lower
limit is 20.
Thus, Lm = 20, N = 153, F = 53, fm = 43, W = 10
⎡ ((N + 1)/2) − (F + 1) ⎤
Median = ⎢ ⎥ W + Lm
⎣ fm ⎦
⎡ (153 + 1) / 2 − (53 + 1) ⎤
= ⎢ ⎥ 10 + 20 = 25.3488 ≈ 25.35%
⎣ 43 ⎦
Thus 25.35% is the median gross profit of the companies.
Example 5.10
A dataset with its class interval and frequency is given below.
Class interval Frequency
0 – 10 15
10 – 20 25
20 – 30 20
30 – 40 X
40 – 50 40
50 – 60 Y
60 – 70 45
It has been given that total number of observations is 210 and the median is 42.5. Find
out the values of missing frequencies X and Y.
Solution:
As the median is 42.5, it lies in 40 – 50 class interval.
(210 / 2) − (60 + X)
42.5 = 40 + × 10
40
⇒ 4 × 2.5 = 105 - 60 - X
⇒ X = 35
The total number of frequency is 210.
∴15 + 25 + 20 + X + 40 + Y + 45 = 210
Putting the value of X,
Y = 30
75
Advantages and Disadvantages of Median

Median is not strongly affected by the extreme or abnormal values. Hence median is a
better average than mean (as seen in example in Table 5.6). Median is easy to
understand and it can be computed from any kind of data (even for grouped data with
open-ended classes, but excluding the case when median falls in the open-ended class).
Median can also be calculated for qualitative data.
However median has some disadvantages. Firstly, it is a time-consuming process as it
is required to arrange the data before calculating the median. Secondly, it is difficult to
compute median for data set with large number of observations. Therefore, to find an
estimate of large population mean is easier to use than median.
THE MODE
Mode is defined as the value of the variable which occurs most frequently in the data
set.
Calculating the Mode from Ungrouped Data

Table 5.13 shows the weights of 20 workers of an organization. The mode is 67 as it
repeats for 4 times (more than other values).
Table 5.13: Weights of 20 Workers
58 60 62 56 59 56 67 68 70 55
67 58 59 60 69 67 67 63 61 70
Calculating the Mode from Grouped Data

When the data is grouped in a frequency distribution the manager must assume that the
mode is located in the class with highest frequency. The mode can be found using the
following equation.
⎛ d1 ⎞
⎜ ⎟
Mode, Mo = Lmo + ⎜ d + d ⎟ w
⎝ 1 2⎠
Where, Lmo = lower limit of the modal class

d1 = frequency of the modal class - the frequency of the class just below it
d2 = frequency of the modal class - the frequency of the class just above it.
w = width of the modal class.
Example 5.11
Consider the salary example in Table 5.10 for computing mode of that data.
Table 5.14: Monthly Incomes of 600 Employees
Class (Rs) Frequency Cumulative Frequency
1000-3000 50 50
3000-5000 110 160
5000-7000 162 322
7000-9000 100 422
76
9000-11000 83 505
11000-13000 45 550
13000-15000 25 575
15000-17000 15 590
17000-19000 8 598
19000-21000 2 600
Solution:
Lmo = Rs. 5000, d1 = 162-100 = 62, d2 = 162-110 = 52 and w = Rs. 2000
⎛ d1 ⎞
⎜ ⎟
Mo = Lmo + ⎜ d + d ⎟ w
⎝ 1 2 ⎠
⎛ 62 ⎞
= 5000 + ⎜ 62 + 52 ⎟ × 2000
⎝ ⎠
= 5000 + 1087.72
= Rs. 6087.72
Advantages and Disadvantages of Mode

Mode can be used as a central location for qualitative as well as quantitative data. It is
not affected by extreme values. It can also be used even when the classes are open
ended.
But mode is not used widely as a measure of central tendency, as it has a few
drawbacks. At times, a data set contains no value that occurs more than once. Further,
all values in a data set might occur at equal number of times i.e. all items are having
the same frequency. Another disadvantage is that some data sets contain two, three or
many modes, making it difficult to interpret them.
Relation between Mean, Median and Mode

In case of a symmetrical distribution, mean, median and mode coincide. However
according to Karl Pearson, if the distribution is moderately asymmetrical, the mean,
median and mode are related in the following manner.
Mean-Median = (Mean-Mode)/3
Thus,
Mode = 3 Median - 2 Mean
Example 5.12
The mean and median for a dataset is 27.5 and 26. Find the mode.
Solution:
We know that,
Mode = 3 Median – 2 Mean
= 3 × 26 – 2 × 27.5
= 23
77
SUMMARY
Central tendency is a statistical measure that takes one number as the representative of
a group. It is an economical estimate of the general characteristics of a group. There
are three main measures: mean, median, & mode. Mean is the average of all the
observations. The median is the middle of a distribution: half the observations are
above the median and half are below the median. The median is less sensitive to
extreme scores than the mean. This makes it a better measure than the mean for a data
set with extreme values. The mode is the most frequently occurring observation in a
distribution.
78
Chapter 6
Measures of Dispersion
• Range
• Interquartile Range and Quartile Deviation
• Mean Deviation
• Variance
• Standard Deviation
• Bienayme Chebyshev’s Rule
The previous chapter has discussed how one can calculate a single value that represents
the characteristics of the entire raw data using three main measures: mean, median, &
mode. Another important characteristic of a data set is how it is distributed, or how far
each element is from some measure of central tendency (average). There are several
ways to measure the variability of the data. Although the most common and most
important is the standard deviation, which provides an average distance for each
element from the mean, several others are also important, and are hence discussed here.
This chapter will discuss various methods to study dispersion like: range, inter quartile
range and quartile deviations, mean deviation, variance and standard deviation.
RANGE
Range is the simplest method of studying dispersion. Range is defined as the
difference between the value of the smallest observation and the value of the largest
observation present in the distribution.
Range = L - S
Where, L = Largest value
S = Smallest value
For a grouped frequency distribution or a continuous frequency distribution, range is
defined as the difference between the upper limit of the highest class and the lower
limit of the smallest class.
Range = Upper limit of the highest class - Lower limit of the lowest class.
Coefficient of Range
The range calculated above is not useful to a manager for comparison if the
observations are in different units. For example, a physical trainer cannot compare the
range of the weights of employees with range of their heights as the range of weights
would be in kilograms and that of heights in centimeters. Therefore, for the purpose of
comparison, a relative measure of range is required, which is called coefficient of
range.
largest value − smallest value
Coefficient of Range =
largest value + smallest value
L −S
=
L+S
Example 6.1
Calculate the range and coefficient of range for the following dataset.
78, 55, 76, 45, 89, 43, 65, 67, 48, 58
Range = L - S
Where, L = Largest value = 89
S = Smallest value = 43
Range = 89 – 43
= 46
largest value − smallest value
Coefficient of Range =
largest value + smallest value
89 − 43
=
89 + 43
= 0.348
80
Example 6.2
Table 6.1 provides the information about the number of employees in an organization
in various age groups. Compute range and coefficient of range.
Table 6.1: Ages of 100 Employees
Class Number of employees
(Yrs) (f)
56-60 7
61-65 16
66-70 39
71-75 28
76-80 10
Total 100
Range = (Upper limit of the highest class) - (Lower limit of the lowest class)
= 80-56 = 24
L −S
And, coefficient of Range = = 24/136 = 0.1765
L+S
Merits and Limitations of Range

Merits:
1 Range is simple to understand and easy to calculate.
2 Range is the quickest way to get a measure of dispersion, although it is not
accurate.
Limitations:
1. It is not based on all the observations in the data. It is computed based on the
highest and the lowest values and ignores the nature of dispersion among other
values of observations in the data set.
2. It is influenced by extreme values and hence fluctuates from sample to sample of a
population, even though the values that fall in between the highest and lowest
values are similar.
3. Range cannot be computed for frequency distributions with open-end classes.
4. Range fails to explain about the character of the distribution within two extreme
observations (i.e. L and S)
5. Range is unreliable as a measure of dispersion of the values within a distribution.
Uses of Range
In spite of the above limitations and short-comings, range, as a measure of dispersion
has many uses.
1 Range is used in industry for the quality control of products without 100%
inspection. Range plays an important role in construction of charts used for quality
control. For example, when the weight of a spare exceeds a particular range the
entire production line is checked to correct the deviation in weight of the spare
produced using range, thus, assuring quality in production process.
2 Range is also useful in studying the fluctuations in financial and share markets.
81
INTERQUARTILE RANGE AND QUARTILE DEVIATION
Range as a measure of dispersion has many limitations as it is based on two extreme

observations. It fails to explain the scatter within the range. So when these extreme
observations are discarded the limited range would be more accurate and representative
of the entire data. This range calculated based on the middle 50 percent of the
observations is called interquartile range. This interquartile range is calculated from
observations obtained after discarding one quartile of the observations at the lower end
and another quartile of the observations at the upper end of the distribution. Thus,
interquartile range is the difference between the third quartile and the first quartile.
The quartiles (Q1, Q2, and Q3) are the highest values in each of the first three of the
four parts of the distribution.
∴ Interquartile range = Q3-Q1
Figure 6.1 shows the concept of interquartile range graphically. Notice that the area
under the curve is divided into four equal parts (25% each).
Figure 6.1: Interquartile Range
Interquartile range
¼ of items ¼ of items
Lowest 1st quartile 2nd quartile 3rd quartile Highest

observation (median) observation
Q1 Q2 Q3
Quartile deviation is defined as one half of the interquartile range.

Q 3 − Q1
Quartile deviation (Q.D.) =
2
Quartile deviation gives the average value by which the two quartiles differ from the
median. In symmetrical distribution, the quartiles Q3 and Q1 are equidistant from the
median i.e.
Median - Q1 = Q3 – Median
This difference can be taken as a measure of variation.
The median ± Quartile deviation covers approximately 50 percent of the observations
as the economic data or any other business data seldom is perfectly symmetrical. A
small quartile deviation denotes less variation in the central 50% of the observations,
where as a high quartile deviation indicates large variations.
82
Coefficient of Quartile Deviation

The relative measure of quartile deviation is called coefficient of quartile deviation. It
can be used to compare the degree of variation in different distributions.
Coefficient of Q.D = Q 3 − Q 1
Q 3 + Q1
Computation of Quartile Deviation

Computation of Q.D is very simple as one needs to calculate the values of the upper
(Q3) and lower (Q1) quartiles.
For Ungrouped Data
Lower quartile (Q1) = (N + 1) t h observation
4
Upper quartile (Q3) = 3(N + 1) t h observation

4
Where, N = total number of observations.
For Grouped Data
Q1 = L 1 ( 14 N − C)
+ ×h
f
Q3 = L 3 ( 34 N − C)
+ ×h
f
Where,
L1 = the lower boundary of the first quartile class (Q1)
L3 = the lower boundary of the third quartile class (Q3)
N = Total cumulative frequency
f = Frequency of the quartile class
h = Class interval (width)
C = Cumulative frequency of the class just above the quartile class
Example 6.3
The following example explains the computation of Quartile Deviation for the data
given in Table 6.2.
Table 6.2 Weights of Employees
Weight (Kg) 55-60 60-65 65-70 70-75 75-80
Number of employees 10 18 14 16 12
Table 6.3: Cumulative Frequency of Weights

Weights Number of employees Cumulative frequency
(frequency) f
55-60 10 10
60-65 18 28
83
65-70 14 42
70-75 16 58
75-80 12 70
Q1 = ¼ N = ¼ (70) = 17.5th observation

This observation will fall in class (60-65)
∴ L1 = 60, C= 10, f = 18, h = 5
Q1 = 60 + 17 . 5 − 10 × 5
18
= 60 + 7 . 5 × 5
18
= 60 + 2.083 = 62.083
Q3 = ¾× 70 = 52.5th observation
It falls in class (70-75)
Q3 = 70 + 52 . 5 − 42 × 5
16
= 70 + 3.28
= 73.28
∴ Quartile Deviation = Q 3 − Q 1 = 73.28 − 62.0823 11.197

= = 5.5985
2 2 2
Coefficient of Q.D. = Q 3 − Q 1 = 73.28 − 62.0823 = 11.197 = 0.0827

Q 3 + Q1 73.28 + 62.0823 135.3623
Merits and Limitations of Quartile Deviation

Quartile deviation (Q.D.) has many merits compared to range and other measures of
variation, but it also has some limitations.
Merits:
1. Q.D can be used as a measure of variation for open-ended distributions.
2. Q.D. is a better measure of variation for highly skewed distribution or distribution
with extreme values as Q.D. is not affected by the presence of extreme values.
Limitations:
1. As the Q.D is calculated using only 50% of the total observations, it cannot be
regarded as a good measure of variation.
2. Q.D. is not a real measure of variation as it does not measure the scatter of
observations from the average. Q.D. is only a positional average.
MEAN DEVIATION
Mean deviation is obtained by calculating the absolute deviation of each observation

from mean.
84
Mean Deviation for Ungrouped Data

To compute mean deviation for ungrouped data, the mean of the sample is calculated.
Then the absolute value of the difference between each item in the distribution (data
set) and the mean is calculated. i.e., subtract the mean from every value in the data set
and ignore the positive or negative signs, (consider everything to be positive). Finally,
all those differences are added and this sum is divided by the number of items in the
sample.
∴ Absolute Mean Deviation = ∑ x − μ (for total population)

N
∑ x−x
Absolute Mean Deviation = (for a sample)
n
Where,
x = value of observation
µ= the mean of population
N = number of observations in the population
x = sample mean
n = number of observations in the sample
Example 6.4
Calculate the absolute mean deviation of the leave patterns of a sample of 10 drivers in
one year for the values given in Table 6.4.
Table 6.4: Calculation of Mean Deviation of the Leave Patterns of 10

Drivers in One Year
Observation in days Deviation from Absolute deviation
S. No. (x−x)
(x) mean (x - x )
(N)
1 10 -11 11
2 15 -6 6
3 18 -3 3
4 20 -1 1
5 20 -1 1
6 22 1 1
7 23 2 2
8 25 4 4
9 27 6 6
10 30 9 9
N=10 ∑x= 210 ∑ x − x = 44
Mean ( x ) = ∑x = 210/10 = 21
N
x −x
∴ Absolute Mean deviation = ∑ = 44/10 = 4.4 days
N
85
Mean Deviation for Grouped Data:

Mean deviation (M.D.) for a grouped or continuous frequency distribution can be
calculated about average (mean) using following equation.
1
M.D. (About the mean x ) = ∑f |x-x |
N
1
= ∑f|d|
N
Where,
x = mid value of the class interval
f = the corresponding frequency
N= total cumulative frequency
| x - x | = the absolute value of the deviations
d = (x- x ) = difference of the values of x from the average x .
Example 6.5
Compute the Mean Deviation for the data given in Table 6.5.
Table 6.5: Classes and their Frequency
Class Interval 0-4 4-8 8-12 12-16
Frequency 4 2 1 3
Table 6.6: Computation of Mean Deviation
Class Frequency Mid-value of f ×X X−X f X−X

Interval class interval
(f)
(X)
0-4 4 2 8 5.2 20.8
4-8 2 6 12 1.2 2.4
8-12 1 10 10 2.8 2.8
12-16 3 14 42 6.8 20.4
N= ∑ f ∑ f × X = 72 ∑ f X − X = 46.4
=10
∑f ×X 72
X= = = 7.2
N 10
1 46.4
Absolute Mean Deviation = ∑f X − X = = 4.64
N 10
86
Merits and Limitations of Absolute Mean Deviation
Merits:
1. Absolute mean deviation is simple and easy to understand.
2. Absolute mean deviation is a more comprehensive measure of dispersion as it is
dependent on all observations of a distribution.
3. As it is obtained by taking the average of the deviations of every observation from
the mean, it is a true measure of dispersion.
Limitations:
1. Absolute Mean deviation is less reliable as it is the arithmetic mean of the absolute
values (ignoring the positive and negative signs).
2. Absolute Mean deviation is not conducive to further algebraic treatment.
3. Absolute Mean deviation cannot be computed for distributions with open-end
classes.
VARIANCE
Variance is similar to mean deviation expect for that, it is calculated using the sum of
the squared distances between the mean and each observation divided by the total
number of elements in the distribution (population). While calculating variance the
differences (deviations) are squared to make them positive.
For Grouped Data
The formula for calculating variance ( σ 2 ) is as follows.
σ2 = ∑ f i (X i − x ) 2
N
For Ungrouped Data

2 2
σ2 = ∑ (X − x )
=
∑ X
− x
2
N N
Where,
σ 2 = Variance
Xi = the value of observation
x = Mean
N = Total cumulative frequency
fi = Frequency of a class
STANDARD DEVIATION
Standard deviation, or σ , is the square root of the average of the squared distances of
the observations from the mean (i.e. square root the variance). The standard deviation
is expressed in same units as those used in the data set.
87
For Ungrouped Data
Standard deviation for ungrouped data ( σ ) = σ2

2
= ∑ (x − x )
N
Where x = the observation

x = the mean
N = ∑ f (total number of observations)
∑ (x - x )2 = the sum of all values of (x - x )2
σ = Standard deviation
σ 2 = Variance
Properties
1. The value of standard deviation remains the same, if in a series each of the
observation is increased or decreased by a constant quantity. In statistical language
we say, standard deviation is independent of change of origin.
For example, for the observations 3, 10 and 12
x = 8.33, σ = 3.859
If we increase the value of each observation by 4.5 we get the observations 7.5, 14.5
and 16.5.
Now x = 12.833
Now σ = 3.859
Hence although x has increased by 4.5, σ remains the same.
2. For a given series, if each observation is multiplied or divided by a constant
quantity standard deviation will also be similarly affected.
Consider the observations 3, 10 and 12. σ = 3.859 as shown in the above
calculation.
Suppose we multiply each observation by 6, the observations become 18, 60 and 72.
x = 50
σ= (18 − 50 ) 2 + ( 60 − 50 ) 2 + ( 72 − 50 ) 2
3
= 23.152.
It is shown that the standard deviation has also been multiplied by 6 i.e.
3.859 × 6=23.152.
Similarly if the observations 3, 10 and 12 with σ = 3.859, are divided by a constant
value, say 2. The observations now become 1.5, 5 and 6.
x = 4.16.
σ= (1 . 5 − 4 . 16 ) 2 + ( 5 − 4 . 16 ) 2 + ( 6 − 4 . 16 ) 2
3
=1.92
Thus the standard deviation is also divided by 2 i.e. 3.859/2=1.92.
88
Therefore, the standard deviation is independent of any change of origin, but is

dependent on the change of scale.
3. Standard deviation is the minimum root-mean- square deviation. In other words,
the sum of the squares of the deviations of items of any series from a value other
than the arithmetic mean would always be greater.
4. As it is possible to compute combined mean of two or more groups, it is also
possible to compute combined standard deviation of two or more groups.
Combined standard deviation denoted by σ12 is computed as follows:
n 1σ 12 + n 2 σ 22 + n 1d 12 + n 2 d 22
= σ 12 =
n1 + n 2
Where,
σ1 = standard deviation of first group
σ 2 = standard deviation of second group
d1 =x 1- x
d2 =x 2- x
x = (n1 x 1 + n2 x 2 ) / (n1 + n2)
Coefficient of Variation
The coefficient of variation is a measure of relative dispersion and is given by
Coefficient of Variation = Standard deviation

Mean
This is generally expressed as percentage
i.e., Coefficient of variation (%) = Standard deviation × 100

Mean
Hence the coefficient of variation measures the spread of a set of data as a proportion
of its mean. It is used in problem situations where we want to compare the variability,
homogeneity, stability, uniformity and consistency of two or more data sets. The data
set for which the coefficient of variation is greater is said to be more variable i.e. less
consistent or less homogeneous. On the other hand, if the coefficient of variation is less
it is said to be less variable i.e., more consistent or more homogeneous.
Example 6.6
Compute the Variance, Standard Deviation and Coefficient of Variation given the
profitability of 50 companies.
Table 6.7: Profitability of 50 Companies

Profit % (xi) Number of companies (fi)
10 15
15 10
20 15
25 6
30 4
89
(10 × 15) + (15 × 10) + (20 × 15) + (25 × 6) + (30 × 4)

x= = 17.4
15 + 10 + 15 + 6 + 4
Table 6.8 Computation of Variance and Coefficient of Variation
xi fi fixi (x i − x) (x i − x)
2
fi (x i − x)
2
10 15 150 -7.4. 54.76 821.40

15 10 150 -2.4 5.76 57.60
20 15 300 2.6 6.76 101.40
25 6 150 7.6 57.76 346.56
30 4 120 12.6 158.76 635.04
Total 50 870 1962.00
x ∑ fixi 870 = 17.4

= =
∑ fi 50
σ 2== ∑ f i (x i − x ) 2 1962 =39.24

=
∑ fi 50
So variance of profitability among 50 companies is 39.24 (Refer to Table 6.6 for

calculation)
Now, Standard Deviation of profits of 50 companies is:
2
S.D ( σ ) = ∑ f i (x i − x )
= 39.24 = 6. 26
N
6.26
Coefficient of variation= Standard deviation = =0.3598=35.98%
Mean 17.4
Example 6.7
A security analyst studied hundred companies and obtained the following Return on
Investment (ROI) data for the year 1992.Calculate the standard deviation in ROI of the
companies.
Table 6.9: ROI of 100 Companies
Returns % 0-10 10-20 20-30 30-40
No. of Companies 19 32 41 8
Solution:
We can find the variability in the ROI of the companies by calculating the standard
deviation for the above data.
The steps involved are:
• Find mean for grouped data.
• Find deviations from mean for grouped data.
• Find square of above deviations.
• Sum up the squared deviations taking frequency into account.
• Take square root.
90
Table 6.10: Calculation of Standard Deviation
Return on No. of
Mid-point Deviation
investment companies
% X f fX X-x f(X- x )2
0-10 5 19 95 -13.8 3618.36

10-20 15 32 480 -3.8 462.08
20-30 25 41 1025 6.2 1576.04
30-40 35 8 280 16.2 2099.52
Total 100 1880 7756.00
Mean µ = ∑ fX 1880
= = 18 . 8 %
∑ f 100
2
Standard Deviation = ∑ f (X − x ) = 7756 = 8.81%
∑ f 100
Thus, the standard deviation for the return on investment is 8.8%.

In this calculation, we always assume that all the observations in a class interval are
located at the mid-point of the class. For example, the first class interval has mid-point
5 and frequency 19. Hence, the assumption is that all the 19 companies have an ROI of
5%.
BIENAYME CHEBYSHEV’S RULE
This rule was developed by Russian mathematicians named Bienayme and P.L.
Chebyshev. It says that what ever may be the shape of a distribution, at least 75 percent
of the values in the population will fall within ± 2 standard deviations of the mean and
at least 89 percent will fall within ± 3 standard deviations from the mean.
The rule states that the percentage of data observations lying within ± k standard
deviations of the mean is atleast ⎛⎜ 1 − 12 ⎞⎟ × 100
⎝ k ⎠
This formula applies to differences greater than one standard deviation about the mean,
and k must be greater than 1.
In case of a symmetrical bell-shaped curve, we can say that:
1. Approximately 68 percent of the observations in the population fall within ±1
standard deviation from the mean
standard deviations from the mean.
standard deviation from the mean.
The diagrammatic representation of the location of observations around the mean of a
bell-shaped frequency distribution is given in Figure 6.2
91
Figure 6.2: Diagrammatic Representation of Bienayme – Chebyshev

Rule for a Bell shaped Curve
99%
95%
68%
μ-3σ μ-2σ μ-σ μ μ+σ μ+2σ μ+3σ
SUMMARY
Measures of dispersion indicate how much the data in a given set of numerical data are
spread out. The simplest of these measures is the range. The range is the distance
between the largest and the smallest value in the data set. Of all measures of
dispersion, the range is most sensitive to extreme values. Similar in spirit to the range,
but not affected by outliers, is the interquartile range, that is, the difference between the
third and first quartiles. In a sense, the interquartile range is the range of the middle
half of the data. The mean deviation and the standard deviation indicate how much the
average value in the data set differs from the mean. Whereas the mean deviation is just
that - the mean distance of the measurements from the mean, the standard deviation is
usually calculated by first finding the variance and then extracting the square root of
the variance. Although more difficult to compute than the other measures of dispersion,
the standard deviation and the variance are the most useful and accurate methods.
92
Chapter 7
Index Numbers
• Definition of Index Number
• Characteristics of Index Numbers
• Uses of Index Numbers
• Types of Index Numbers
• Problems Related to Index Numbers
• Methods of Constructing Index Numbers
• Quantity and Value Indices
In organizations, managers generally face many business related problems, where in

they are interested in transforming the data into purely relative numbers, which can be
used for comparison. In the earlier chapters we have discussed how averages are used
to find the central values of a distribution and also studied the average of deviations
from the central value. This chapter will focus on ways of studying a group of related
variables over a period of time using index numbers. Index numbers are used for
comparing quantitative data relating to production, consumption, profits, personnel and
finance.
For example, a manager may use price index (measures overall price changes of a
variety of goods and services over a period of time) to construct a measure of the
change in material cost in a 20-year period. In India the salary and wages of employees
and workers are tied to the cost of living index and their salary will increase or
decrease for every change of 8 or 12 or 16 points in the cost of living index. This
chapter discusses the concept of index numbers and explains various types of index
numbers and their use. This chapter also discusses various methods of constructing
index numbers.
DEFINITION OF INDEX NUMBERS
An index number is a statistical measure designed to show the changes in variables or a

group of related variables with respect to time, geographic locations or other
characteristics of the variable, under study.
Price index, quantity index and value index are some examples of index numbers.
Usually these index numbers are calculated by finding the ratio of the current value to
a base value. This ratio multiplied by 100 gives the index in the form of a percentage.
This final value is called percentage relative.
CHARACTERISTICS OF INDEX NUMBERS
Index numbers are defined by different people in different ways. They are referred to
as measures of change, a device to measure change or a series representing the process
of change. Depending on these definitions, the following are some of the
characteristics of index numbers.
Modified Form of Average

Index number is a special form of average as it is generally a weighted average derived
from a sample of items identified to be representative of the entire universe
(population). It is considered to be a special type of average as these average variables
are expressed in different units of measurement.
Percentage Indicating Relative Change

Index numbers can be expressed in percentages as they show the extent of relative
change. But the sign % is not used in index numbers.
Comparative Nature
Index numbers by nature act as devices to measure change and hence they can be used
to compare two variables. They compare changes taking place over time among
variables in a category.
94
Index Numbers
USES OF INDEX NUMBERS
In the beginning, index numbers were constructed to study the price levels over a
period of time. But today these index numbers are extensively used in fields of
economics and business management to compare data relating to production, sales,
revenues and other financial matters. Some of the important uses of index numbers are
as follows.
Barometer of Economic Activity

The first and the most important use of index numbers is that they are used as devices
to measure the level of economic activities.
Measures Comparative Changes in Economic Phenomena

The second use of index numbers is that they are used to measure cross-sectional
changes over a base figure. The changes in the economic phenomena like price level
and cost of living cannot be measured directly and index numbers act as the devices for
measurement as they reduce the complexity of measurement and show the change in
variables over a period of time. The measurement taken at different times can be
readily compared with a base value.
Provide Framework for Decision Making

Index numbers play a major role in the decision making process of the government and
business firms as the index numbers like whole sale and retail price indices, output
index numbers (related to volume of trade, industrial and agricultural production) lay
guidelines for business policy-making.
Forecast Future Events

Index numbers study the relative changes in the levels of economic phenomena at
different points of time and hence they are helpful in studying the trend in a time series
data. Thus, index numbers can be used to forecast the future events.
Measure Purchasing Power of Money

The real wages can be obtained by dividing the wages by the price index and
multiplying the resulting number by 100. This real wage in turn forms a basis for
determining the purchasing power of money.
Calculating Real National Income

Index numbers are used to adjust the net national income against inflation. The
adjusted net national income with respect to present prices or constant prices of a given
base year is known as net national product. This is used to measure the real income.
TYPES OF INDEX NUMBERS

At present there are many kinds of index numbers that are being used across the globe
depending on their purpose. The three important index numbers used are:
• Price index
• Quantity index
• Value index
95
Price Index
Price index compares the changes in prices of a basket of commodities from one period
to another. General price index is used to measure the value of money. Price index
numbers are the most popularly used devices employed in various business and
economic decisions. This price index is further classified into wholesale price index
numbers and retail price index numbers.
Wholesale price index numbers
These index numbers give the change in the general price levels in a country. Official
wholesale price index in India was first complied by the Economic Advisor (Ministry
of Commerce) in 1947 (base year August 1939).
Retail price index numbers
Retail price index numbers give the changes in the retail prices of various commodities
like consumables, stocks, shares, bank deposits, government bonds, etc. Labor Ministry
prepares these retail price index numbers separately for urban and rural areas.
Cost of living index is a type of retail price index, which helps in measuring the impact
of changes in price of selected basket of goods on the purchasing power of people of
various income groups.
Quantity Index
Quantity index number measures how much the number or quantity of a variable
changes over time. For example, it measures the changes in the volume of goods
produced, purchased or consumed, the indices of agricultural production, industrial
production, imports and exports, etc. To be precise, quantity index numbers help in
comparing the level of physical output in an economy over a period of time.
Value Index
Value index measures the change in total monetary value over a period of time. Value
index combines price and quantity changes; here both the total value of the product and
value added is taken. Hence, it is useful in cases like sales, inventory, foreign trade,
etc.
PROBLEMS RELATED TO INDEX NUMBERS
Though index numbers offer several advantages, there are many problems which will
influence the construction of index numbers. Some of the problems are as follows:
Inadequate Data
Sometimes, managers find it difficult to obtain suitable data for computing an index.
For example, suppose the sales manager of a fertilizer company want to develop an
index that would describe the seasonal variations in the fertilizer sales. If the sales
figures are available only on yearly basis, he cannot design an index to measure the
seasonal sales pattern.
Incomparability of Indices
This problem arises when one tries to compare one index with another when there is a
change in the basis product or service that is being measured. For example, when we
compare price indices of automobiles from 1990 to 2000, we find that prices have
96
Index Numbers
increased substantially. But this comparison fails to take into consideration the
technological advances in the design and the quality of automobiles manufactured over
a time period (1990-2000).
System of Weights
In developing a composite index like consumer price index all commodities included
are not of equal importance. Hence, the system of weights and the allocation of weights
to different commodities is very important. Inappropriate weighting of commodities
can distort the index. For example, a rupee per liter increase in price of diesel cannot be
compensated by a rupee decrease in road tax as the increase in price of diesel has a
much greater impact on the consumer than that of decrease in the road tax.
Choice of Base or Reference Period

The major problem faced while studying a phenomenon is that of selecting a base
period of normal and stable economic conditions. For example, when a firm selects a
year of highest profit as its base, then its present profits will obviously get decreased
and vice versa. Managers should take care that the base period is not too distant from
the present.
METHODS OF CONSTRUCTING INDEX NUMBERS

For developing the above discussed index numbers, various methods can be used for
computation. These methods can be grouped under two forms, the Aggregative method
and Average of Relatives methods. In aggregative method the index numbers are
computed based on the aggregates of values of groups of variables for the base period
and the current period. But in the relatives method this comparison is based on the
“relatives”. Relatives are the ratios of the values for the current period over that of the
base period. Firstly, the values of “relatives” are calculated, and later the index
numbers are computed by taking the average of these relatives.
Unweighted Aggregates Index

Unweighted aggregates index is the simplest form of a composite index. The term
“unweighted” implies that all the items (or) values considered in computing the index
are given equal importance. “Aggregate” refers to a sum of values.
In this method, index number is calculated by adding all the elements (price, quantity,
etc.) for the given time period and dividing this sum with the sum of same elements
during the base period. In other words, the sum of all the elements is expressed as a
percentage of the same in the base year. The formula for computing unweighted price
and quantity index are given below:
Unweighted aggregates price index = ∑ P1 × 100
∑ P0
Where
P1 = Price of each element during the current year
P0 = Price of each element during the base year
Unweighted aggregates quantity index = ∑ Q 1 × 100
∑ Q0
Where
Q1= Quantity of each element during the current year
Q0= Quantity of each element during the base year.
97
Suppose, we are calculating price index using unweighted aggregate method, we use
price as a variable and quantity as weight. The following example (Table 7.1) shows
the computation of price index using unweighted aggregative method.
Table 7.1: Calculation of Price Index Using Unweighted Aggregates
Method
PRICE
Commodities 1990 (in Rs) 2000(in Rs)
A 30.00 40.00
B 20.00 25.00
C 6.00 12.00
D 1.00 1.50
∑ P0 = 57.00 ∑ P1= 78.50
Then unweighted aggregate price index is
= ∑ 1 × 100 = 137.7
P
∑ P0
So the unweighted aggregate price index is 137.7, it implies the prices of commodities
has increased by 37.7 percent from 1990 to 2000.
The major disadvantage of this method is that it fails to assign greater weight to high-
consumption good over low-use goods. Hence it fails to be a real measure of change in
prices over a period of time.
Weighted Aggregates Index

This method assigns weights to various commodities depending on their relative
importance in the group. Defining a weight is the most important step in the process of
developing index numbers using this method.
If ‘Q’ is the weight assigned to a commodity, then the formula of calculating a
weighted aggregates price index is:
Weighted aggregates price index = ∑ Pi Q × 100
∑ P0 Q
Where,
Pi = Price of each commodity in the group in the current year
P0 = Price of each commodity in the group in the base year
Q = Weight attached to the commodity
The following example (shown in Table 7.2) explains the computation of price index
using weighted aggregates index.
Table 7.2: Computation of Price Index Using Weighted Aggregates Index
Price P0Q Pi Q
Commodity Quantity
1990 2000
A 10 (Kg) 300 400 3000 4000
B 5 (Litres) 100 150 500 750
C 2 Units 50 100 100 200
∑ P0Q = 3,600 ∑ PiQ = 4,950
98
Index Numbers
Weighted aggregates index is

∑ Pi Q
= × 100 = 4950/3600 × 100 = 137.5
∑ P0 Q
Thus the weighted aggregates price index is 137.5.
Methods to weight an index
Generally, quantity of a commodity consumed is used as a measure of its importance in
computation of a weighted aggregate index. But one faces problems in selecting a
weight. There are four methods commonly used to weight an index, viz.
• Laspeyres Method
• Paasche Method
• Fisher Method
• Fixed Weight Aggregates Method
Laspeyres method
This method uses the quantities used or consumed in the base period in computing each
index number. This method is named after a German economist Etienne Laspeyres
who developed it in 1871. This is the most commonly used method because it requires
quantity measures of only one period (since each index number is developed based on
the same base price and quantity). This enables the managers to compare the index of
one period directly with that of another period. While developing the Laspeyers price
index the product of the current period prices (Pi) and the base periods quantity (Q0)
for each commodity in the group (i.e. product = Pi Q0) is calculated and their sums are
found (∑PiQ0). In the second step, the product of base period prices (P0) and the base
period quantities (Q0) are found and the sum (∑P0Q0) of these values are calculated.
Now by dividing the first sum by the second sum (i.e. ∑PiQ0 ÷ ∑P0Q0) and multiplying
the resulting value by 100 we can convert this value into percentage relative. The
following is the formula for computing the Laspeyres price index.
Laspeyres price index = ∑ Pi Q 0 × 100
∑ P0 Q 0
Where,
Pi = Prices of commodities in the current year
P0 = Prices of commodities in the base year
Q0 = Quantities consumed in the base year
The following example (Table 7.3) shows how to compute a Laspeyres index using the
above formula.
Table 7.3: Calculating Laspeyres Index

Base price Current price Base Quantity
Commodity P0Q0 PiQ0
(P0) (Pi) (Q0)
A 30 40 100 3000 4000
B 20 25 50 1000 1250
C 6 12 20 120 240
TOTAL 4,120 5,490
Laspeyres price index = ∑ Pi Q 0 × 100 = (5490/4120) x 100 = 133.25

∑ P0 Q 0
99
Thus, Laspeyres price index = 133.2. This implies that the prices have increased by
33.2 percent from that of the base year to the present year. The major advantage of
using this Laspeyres index is that one can directly compare the value of any two years
computed against the same base year. The basic disadvantage Laspeyres method is that
it does not consider the change in quantities (consumption) patterns between the base
year and the current year. This will distort the price index.
Paasche method
Paasche method uses the quantity measures for the current period rather than for the
base period (as used in Laspeyres method). Paasche index is calculated similar to the
Laspeyres price index. The formula for calculating Paasche price index is as follows.
Paasche price index = ∑ Pi Q i × 100
∑ P0 Q i
Where
Pi = Prices of commodities in the current period
P0 = Prices of commodities in the base period
Qi= Quantities in the current period
The most important advantage of Passche method is it uses the changes in price and
consumption patterns. Thus, it is a better measure of general changes in the economy
than the Laspeyres’ method of calculating weighted aggregates index. The following
example (Table 7.4) shows how to calculate a Paasche index.
Table 7.4: Calculating Paasche Index
Current Base price Current
Commodity PiQi P0Qi
price Pi P0 Quantity Qi
A 40 30 120 4,800 3,600
B 25 20 100 2,500 2,000
C 12 6 20 240 120
Total 77 56 240 7,540 5,720
∑ Pi Q i
Paasche price index = × 100 = (7540/5720) x 100 = 131.81
∑ P0 Q i
Hence, Paasche price index is 131.81. This implies that the prices have increased by
31.81 percent from the basic year to the current year. Thus each value used in
developing the Paasche price index is the combination of price and quantity changes
from the base year. The primary disadvantage of this method is that it is impossible to
compare the indices from different periods as the quantity measure used for one period
is different from that used in other period. Although Paasche method is a better
indicator of general changes in economy, it cannot be used to directly compare indices
(as possible in Laspeyres method).
Fisher method
Fisher price index number is given by the geometric mean of Laspeyres and Paasche
formula. The formula for computing Fisher’s index is as follows.
Fisher Index = Laspeyres index × Paasche index
∑ Pi Q 0 ∑ Pi Q i
= × 100 × × 100
∑ P0 Q 0 ∑ P0 Q i
100
Index Numbers
Hence in the above example, Fisher index is
= 133.25 ×131.81 = 132.52
Fixed weight aggregates method

This method is similar to Laspeyres and Paasche methods. However, instead of using
base period or current period quantities (weight), this method uses weights from a
model (representative) period. The weights of this period are known as fixed weights.
These fixed weights and the base prices need not belong to the same period.
The fixed weight aggregates index is computed by multiplying the current period
prices by the fixed weights and adding the results. Then the base period price is
multiplied by the fixed weights and the results are added. Now the first sum is divided
by the second and multiplied by 100 to get the ratio in percentage form.
The formula for calculating a fixed weight aggregates price index is as follows.
Fixed weight aggregates price index =
∑ Pi Q 2
= × 100
∑ P0 Q 2
Pi= Prices of commodities in the current period
P0= Prices of commodities in the base period
Q2= Fixed weights
Table 7.5 shows the computation of a fixed weight aggregates price index.
Table 7.5: Computation of a Fixed Weight Aggregates Price Index
(1990) (2000)
(1995) (1990) (2000)
Weight Weight
Commodity Quantity Average Average
aggregate Aggregate
(Q2) price (P0) price (Pi)
(P0Q2) (PiQ2)
A 120 30 40 3600 4800
B 70 20 25 1400 1750
C 20 6 12 120 240
Total ∑P0Q2 = 5,120 ∑P1Q2 = 6790
∑ Pi Q 2
Fixed weight aggregates price index = × 100 = (6790/5120) × 100 = 132.62
∑ P0 Q 2
Hence fixed weight aggregates price index is 132.6. The basic advantage using this
method is that it gives the manager the freedom to select the base price and the fixed
quantities (capital). This allows the manager to select the base price and quantity from
different periods to develop more accurate indices.
Unweighted Average of Relatives Method

In average of relatives method, first the ratio of the current price to the base price for
each product is calculated and each of these ratios is multiplied by 100 to get the
percentage relatives. Then the sum of these percentage relatives is divided by the
101
number of products (commodities). The formula for the unweighted average of

relatives method is as follows:
⎛P ⎞
∑ ⎜⎜ i × 100 ⎟⎟
P
Unweighted average of relatives index = ⎝ 0 ⎠
n
Where
Pi= Price of commodities in the current period
P0= Price of commodities in the base period
n = Number of commodities in the group
Table 7.6: Computation of an Unweighted Average of Relatives Index
PRICE (Rs.) Ratio × 100

Commodity Ratio (Pi/P0)
1990 (P0) 2000 (Pi) (Pi/P0) × 100
A 30.00 40.00 40/30 = 1.33 1.33× 100 ≅133
B 20.00 25.00 25/20 = 1.25 1.25× 100 = 125
C 6.00 12.00 12/6 = 2.00 2.00 × 100 = 200
D 1.00 1.50 1.5/1 = 1.5 1.5 × 100 = 150
⎛P ⎞
∑ ⎜⎜ i × 100 ⎟⎟
P
Unweighted average of relatives index = ⎝ 0 ⎠ = 608/4 = 152
n
From this calculation it is clear that the general price index for year 2000 is 152 using
unweighted average of relatives index. But, for the same problem when the index was
calculated by using unweighted aggregates it was 137.7. The difference in indices
calculated by these two methods arises due to the fact that in unweighted average of
relatives method the average of the ratios of the prices for each commodity is
computed separately, and with the unweighted aggregates method the ratio of sums of
the prices of each product is calculated. Instead of assigning weights, the unweighted
average of relatives’ method converts each commodity into a relative scale where each
of them is shown as a percentage.
Weighted Average of Relatives Method

To enhance the accuracy of the index this method uses weights to develop the index.
While using this weighted average of relatives method a manager has to use the value
of each commodity in the group (the value is the total monetary (Rs.) volume obtained
by multiplying price by quantity).
For this method, the weight values can be calculated using any of the weighting
methods (Laspeyres method, Paasche method etc). By using Laspeyres method, the
base value can be computed by multiplying the base quantity by the base price. Thus,
the index calculated by weighted average of relatives method based on base value is
the same as that calculated using Laspeyres method.
102
Index Numbers
Table 7.7: Computing a Weighted Average of Relative Index

Price Weighted
Quantity Base
Relative Pi/P0 × 100 Percentage
Commodity 1990 2000 1990 Value
(1) Relative =
(P0) ( Pi ) (Q0) P0Q0 (2)
(1) × (2)
A 30 40 100 40/30 × 100 = 133.33 3000 ≅ 4,00,000

B 20 25 50 25/20 × 100 = 125 1000 125,000
C 6 12 20 12/6 ×100 = 200 120 24,000
∑P0Q0 =
Total 5,49,000
4,120
The formula used to calculate the weighted average of relatives method is as follows:
Weighted average of relatives price index (General form)
⎡⎛ P ⎞ ⎤
∑ ⎢ ⎜⎜ i × 100 ⎟⎟ ( P n Q n ) ⎥
P
= ⎣⎢ ⎝ 0 ⎠ ⎦⎥
∑ Pn Q n
Where,
PnQn= The base values
P0=Prices of commodities in the base period
Pn= Prices for base period (or) current period (or) fixed period.
When weighted average of relatives index is computed using the base values, the
equation would be as follows:
Weighted average of relatives price index (using base values)
⎡⎛ P ⎞ ⎤
∑ ⎢ ⎜⎜ i × 100 ⎟⎟ ( P 0 Q 0 ) ⎥
= ⎢⎣ ⎝ P 0 ⎠ ⎥⎦
∑ P0 Q 0
The value of index computed using the above equation would produce the same results
for any problem as that obtained by using the Laspeyres method. The following
example shows the computation of price index using weighted average of relatives’
method.
Weighted average of relatives index
⎡⎛ P ⎞ ⎤
∑ ⎢ ⎜⎜ i × 100 ⎟⎟ ( Pn Q n ) ⎥
P
= ⎢⎣ ⎝ 0 ⎠ ⎥⎦ = ⎡ 549000 ⎤ = 133.25
⎢ 4120 ⎥
∑ n n
P Q ⎣ ⎦
Weighted average of relatives index is 133.25.
QUANTITY AND VALUE INDICES

In this chapter, all the methods of developing index numbers were explained using the
price index. But index numbers can also be used to describe the value and quantity
changes. The formula for weighted average of quantity index is as follows
⎡⎛ Q ⎞ ⎤
∑ ⎢⎜⎜ i × 100 ⎟⎟(Q n Pn ) ⎥
⎢ Q ⎦⎥
Weighted average of relatives quantity index = ⎣⎝ 0 ⎠
∑ Q n Pn
103
Where
Qi= Quantities of commodities used in the current period
Q0=Quantities of commodities used in the base period
Qn and Pn =Quantity and price values used as weights
The major advantage of using the quantity index is that it gives a better measure of the
changes in the economy in times of inflation because it measures the actual output of
raw materials and finished goods rather than taking the price of goods.
Value index measures general changes in the total value of some variable. It measures
the combined effects of price and quantity changes. The index numbers can also be
used to describe the changes in quantity and value.
Example 7.1
The consumption of material by a bakery is given below.
Price (in) Quantities used
Inputs Units Po P1 Qo Q1
1993 1996 1993 1996
Flour Kilo 15 30 500 700
Eggs Dozen 8 14 100 70
Milk Litre 6 17 200 120
Sugar Kilo 10 16 50 70
Calculate Laspeyre’s price index, Laspeyre’s quantity index, Paasche’s quantity index
and unweighted average of relative index.
a. Laspeyre’s price Index for the given data is
P1Q0 P0Q0 P0Q1 P1Q1 ⎛ P1 ⎞
⎜⎜ ⎟⎟ × 100
⎝ P0 ⎠
15000 7500 10500 21000 200

1400 800 560 980 175
3400 1200 720 2040 283.3333
800 500 700 1120 160
∑ P1Q0 = ⎛P ⎞
∑ P0Q0 ∑ ⎜⎜ 1 ⎟⎟ × 100 = 818.3333
20600 =10000 ∑ P0Q1 =12480 ∑ P1Q1 = 25140 ⎝ P0 ⎠
Laspeyre’s price index = ∑ P1 Q 0 × 100
∑ P0 Q 0
20600
= × 100
10000
= 206
b. Laspeyre’s quantity index for the given data is
Laspeyre’s quantity index = ∑ Q 1 P0 × 100
∑ Q 0 P0
12480
= × 100
10000
= 124.8
104
Index Numbers
c. Paasche’s quantity index for the given data is

Paasche quantity index = ∑ 1 1 × 100
Q P
∑ Q 0 P1
25140
= × 100
20600
= 122.0388
d. The unweighted average of relatives price index for the given data is
⎛P ⎞
∑ ⎜⎜ 1 × 100 ⎟⎟
P0
Unweighted average of relatives index = ⎝ ⎠
n
818.333
=
4
= 204.5833
SUMMARY
An index number is a statistical measure designed to show changes in variables or a
group of related variables with respect to time, geographic locations or other
characteristics of the variable under study. It is referred to as a measure of change, a
device to measure change or a series representing the process of change. Index
numbers are used as a barometer to indicate the changes in economic activity. They
also provide framework for decision making and to forecast future events. There are
three types of index numbers which are generally used. They are price index, quantity
index and value index. These index numbers can be developed either by aggregate
method or by average of relatives method. This chapter discussed various methods of
weighting an index like the Laspeyres method, Paasche method, Fisher method and the
fixed weight aggregates method. Although there are many problems related to
developing an index it is a handy device to measure or compare the changes in
economic variables over a period of time.
105
Chapter 8
Introduction to Probability
and Probability
Distributions
• Basic Probability Concepts
• Types of Probability
• Probability Rules
• Random Variables
• Types of Probability Distributions
• Normal Distribution
Intro to Prob. and Prob. Distributions
The concept of probability originated in the seventeenth century and has become one
of the most fascinating and debatable subjects in the recent years. Probability has
gained a lot of importance and the mathematical theory of probability has become the
basis for statistical applications in the areas of management, space technology and the
like. In fact, most of the people use probability in their day-to-day lives without being
aware of it. Statements like “It may rain today”, “Probably I will continue with the
same job”, “India might win the cricket series against Australia,” etc. are examples of
the usage of probability in day-to-day life.
Various business decisions in real life are made under situations when a decision
maker is very uncertain as to what will happen after the decisions are made. The theory
of probability is of great help in all such areas. In particular, it enables a person to
make ‘educated guesses’ on matters where either full facts are not known or there is
uncertainty about the outcome. The probability formulae and techniques were
developed by Jacob Bernoulli, De Moiure, Thomas Bayes, and Joseph Lagrange. Later
on Pierre Simon and Laplace unified all these early ideas and compiled the first general
theory of probability. Even though, volumes have been written on probability, the
controversies concerned with the concepts of probability theory continue.
The concept of probability was used by gamblers during the early days in games of
chance such as throwing a die, drawing a card from the deck or tossing a coin. In these
games of chance, there is an uncertainty regarding the face of the die that will appear in
a throw or the card that will appear in a draw or the face of a coin that will appear
when it is tossed. Although there is an uncertainty concerning the outcome of any
particular throw or any particular drawing, there is a predictable long-term outcome.
For instance, if a die is thrown many times, experimental studies have shown that the
probability of a number to appear is one sixth (as the die has 6 faces).
BASIC PROBABILITY CONCEPTS
Before defining the term ‘probability’, it is necessary to familiarize with certain terms
that are used in this context.
Experiment
Any operation / process that results in two or more outcomes is called an experiment.
Examples of an experiment:
• Rolling an unbiased die is an experiment, where the number that is to appear on
the face of the die is unpredictable and subject to change.
• Tossing a fair coin is an experiment, where the outcome head or tail is
unpredictable.
Random Experiment
Any well-defined process of observing a given chance phenomena through a series of
trials that are finite or infinite and each of which leads to a single outcome is known as
a random experiment.
Examples of random experiment:
• Drawing a card from a pack of 52 cards. This is also a chance phenomenon with
only one outcome.
• Drawing a ball from a bag containing a given number of red, blue and white balls.
This is also a chance phenomenon with only one outcome.
107
A random experiment is different from experiments under control conditions (example,

experiment in a physical laboratory) because the observation in a random experiment
involves chance phenomena and is not performed under controlled conditions.
Possible Outcome
The result of a random experiment is called an outcome. For example, picking a card
from a pack of 52 cards and getting an ace or a Jack or a Queen or a King is an
outcome.
Event
An event is one or more possible outcomes of an experiment or a result of a trial or an
observation. In other words, an event is used to denote a phenomenon that occurs with
every realization of a set of conditions.
Elementary Event
A simple or elementary event is a single possible outcome of an experiment. A simple
event cannot be further subdivided into a combination of other events.
Example
Throwing a die and the event of getting a six (6) is a simple event.
Compound Event
When two or more events occur in connection with each other, then their simultaneous
occurrence is called a compound event. The compound event is an aggregate of simple
events.
Example
When we roll two dice, then the event of getting a six on either the first or second die is
a compound event.
Favorable Event
The number of cases favorable to an event in a trial is the number of outcomes that
result in the happening of a particular event.
Examples
• In drawing a card from a pack of 52 cards, the number of favorable cases for
drawing an ace is 4, for drawing a spade are 13 and for drawing a black card are
26.
• In throwing of three die, the number of cases favorable to getting the sum of 4 is:
(1, 1, 2), (1, 2, 1), (2, 1, 1), i.e. totally three favorable outcomes.
Mutually Exclusive Events

Two events are said to be mutually exclusive or incompatible if the happening of any
one of them precludes the happening of all others i.e., both the events cannot happen
simultaneously in a single trial or, the happening of one prevents the happening of the
other and vice-versa.
Examples
• In throwing a die, the events of getting each of the six faces numbered 1 to 6 are
mutually exclusive since if any one of these faces comes, the possibility of others,
in the same trial is ruled out.
• If a single coin is tossed, head can be up or tail can be up, both cannot be up at the
same time.
108
Dependent or Independent Events

Two or more events are said to be independent if the happening of an event is not
affected by the supplementary knowledge concerning the occurrence of any number of
the remaining events. The question of dependence or independence of events is
relevant when experiments are consecutive and not simultaneous.
Examples
• In tossing an unbiased coin, a trial is not affected by the result of the previous of
subsequent trails. The events therefore are independent.
• If a card is drawn from a pack of 52 well-shuffled cards, then only 51 cards are
left. Now, if a second card is drawn by replacing the first card (the picked card)
then the pack again has 52 cards and the trials are independent. But, if the first
card is not replaced back, the composition of the pack stands changed and the
probability of the second card is affected and thus the event is dependent on the
previous trial.
Exhaustive Events
The total number of possible outcomes in any trial is known as exhaustive events or
exhaustive cases.
Examples
• In tossing a fair coin, there are two possible outcomes, head and tail. The list of
these outcomes is exhaustive since the result of any toss must be either head or
tail, if the possibility of the coin standing on an edge is ignored.
• For throwing two dice, the exhaustive number of cases is 62 = 36. In general, for
throwing ‘n’ dice, the exhaustive number of events is 6n. This is because any of the
six numbers from 1 to 6 of the first die may be associated with any of the six
numbers of the other dice.
The sum of the probabilities for a mutually exclusive event or an exhaustive event
should be equal to one.
Equally Likely Events

Events are said to be equally likely, if taking into consideration all the relevant
evidence, there is no reason to expect one in preference to the others. In other words,
when an event does not occur more often than the others, they are said to be equally
likely events.
Examples
• In throwing an unbiased die, the outcome of a number from 1 to 6 is equally
likely.
• In picking a card from a pack of 52 cards (with replacement), each card can be
picked up equally often.
• When an unbiased coin is tossed, the event of getting either head or tail is equal.
Complementary Events
A complementary event is the number of unfavorable cases in an experiment. Suppose
‘E’ is an event of the number of favorable cases in the experiment, then a
complementary event denoted by E is the number of unfavorable cases in that
experiment. The events E and E are mutually exclusive and exhaustive.
109
Examples
• In drawing a card from a pack of 52 cards, the event of getting an ace of diamond
is only one and that of getting the complementary i.e., unfavorable event is 51.
• In throwing a die, the favorable event of getting a face with number 1 is 1 and the
unfavorable event of getting it is 5.
The sum of the probabilities of an event and its complementary event is one.
TYPES OF PROBABILITY
There are three basic ways of classifying probability based on the conceptual
approaches to the study of probability theory. There is disagreement among the experts
regarding the appropriate approach of probability. The basic approaches are:
• Classical approach
• Relative frequency of occurrence approach
• Subjective approach
Classical Approach
The classical approach is based on the assumption that each event is equally likely to
occur. This is an apriori assumption (the term apriori refers to something that is
known by reason alone) and the probability based on this assumption is known as
apriori probability. This approach employs abstract mathematical logic and hence is
also called as ‘abstract’ or ‘mathematical’ probability. This is the reason for
considerable use of familiar objects like cards, coins, dice, etc., where the answer can
be stated in advance before picking a card, tossing a coin or throwing a die,
respectively.
Definition
If a random experiment results in ‘N’ exhaustive, mutually exclusive, and equally
likely outcomes, out of which ‘f’ are favorable to the happening of an event ‘E’, then
the probability of occurrence of E, usually denoted by P(E) is given by
P = P (E) = f / N
This is rather a complex way of defining something that may seem intuitively obvious,
but it can be used to tossing a coin and throwing a die example.
(or)
Probability of happening an event (E) = Number of favourable outcomes

Total number of outcomes
James Bernoulli was the first man to obtain a quantitative measure of uncertainty and
the above definition was given by him. The above formula can also be expressed as
‘the odds in favor of E’ are f: N or ‘the odds against E’ are (N-f): N.’
The probability ‘q’ that the event ‘E’ will not happen is given by
N−f N f f
q = Q (E) = = − = 1− = 1− p
N N N N
⇒ p + q= 1 (as q = 1 - p)
(or)
The probability of not happening an event = 1- Probability of happening an event.
110
Hence, the probability of happening of an event plus the probability of not happening
of the event is always equal to unity. Obviously, p as well as q are non-negative and
cannot exceed unity i.e., 0 ≤ p ≤ 1 and 0 ≤ q ≤ 1.
If the probability of an event, P (E) is zero then, the event is called an impossible event
and if P (E) = 1 then the event is called a certain event.
Classical approach can be illustrated for tossing of a coin or a die. Suppose that the
probability of getting a head on a single toss is to be calculated, then using formal
terms,
P (Head) = 1 = 0.5
1+1
The probability of getting ‘3’ on a single throw of a die is to be calculated, then using
formal terms,
P (3) = 1 = 1/6 = 0.167
1+1+1+1+1+1
Example 8.1
What is the probability of drawing a face card (King, Queen or Jack) from a pack of
well shuffled pack of 52 cards?
Solution:
The total number of face cards in a pack of 52 = 12 cards
P (Drawing a face card) =

(Total number of face cards)
(Total number of cards)
12 3
= =
52 13
Example 8.2
If two fair dice are tossed simultaneously, what is the probability of getting same
number on both the dice?
Solution:
If two dice are tossed simultaneously, the sample space would hold 36 combinations.
There will be six combinations where both the numbers in a combination will be same.
Therefore, the probability of getting the same number on both the dice is 1/6.
Limitations of classical approach to probability
The classical approach to probability assumes a world that does not exist or is highly
hypothetical in its assumptions. It assumes situations that are very unlikely but could
conceivably happen. The limitations of this approach are:
1. The classical definition is applicable only when the trials are equally likely or
equally probable. For instance, the probability that a candidate, attending an
interview, will succeed is not 50% since the two possible outcomes viz. success
and failure are not equally likely.
2. The classical definition is applicable only when the exhaustive number of cases in
a trial is finite.
3. The classical definition is applicable only when the events are mutually exclusive.
Thus, the classical approach to probability is useful in card games, dice games, tossing
coins and the like, but has serious problems when it is applied to less orderly decision
problems that are encountered in the area of management. Probabilities of occurrences
111
such as an employee resigning from a job before his/her retirement age or the delay in
delivery of a product to a nearby customer cannot be predicted using this approach.
Relative Frequency of Occurrence Approach

The relative frequency of occurrence approach defines the probabilities as either:
• The proportion of times that an event occurs in the long-run when the conditions
are stable, or
• The observed relative frequency of an event in a very large number of trials.
In this approach, the probability of happening of an event is calculated knowing how
often the event has happened in the past. In other words, this method uses the relative
frequencies of past occurrences as probabilities. For instance, suppose that an
organization knows from the past data that about 25 of its 300 employees entering
every year leave the organization due to good opportunities elsewhere. Then the
organization can predict the probability of the employee turnover for this reason as
25 / 300 = 1/12 = 0.083
Another characteristic of probabilities established by the relative frequency of
occurrence approach can be illustrated by tossing a fair coin 1000 times. In this case it
is found that the proportion of getting either a head or tail is more initially but as the
number of tosses increase, both a head or tail become equally likely and the probability
of the event showing a head is 0.5 or the event showing a tail is 0.5. Thus, accuracy is
gained as the experiment is repeated and the number of observations is more. But the
limitation of this approach is the consumption of time and cost for such large
repetitions and additional observations. Moreover, predicting probability using this
approach becomes a blunder if the prediction is not based on sufficient data.
Subjective Approach
Subjective probabilities are those assigned to events by the manager or the researcher
based on the past experiences or occurrences or on the evidences available. It may be
an educated guess or intuition. At higher levels of managerial decisions, when the
decision making becomes very important, specific and is demanded to be unique,
managers use subjective probability. The approach was introduced by Frank Ramsey in
1926.
PROBABILITY RULES
A probability is a number assigned to the occurrence of an event in a sample space.

Probability measures must satisfy three rules. If A is an event with probability denoted
by P (A), then the following rules hold:
Rule 1
The probability of the entire sample space S is 1, i.e. P(S) = 1.
Area of sample space rectangle = 1. An event A is represented within the rectangle.
Minimum possible area of A is 0 and maximum possible area is 1.
Rule 2
The probability of the event A must be greater than or equal to 0 and less than or equal
to 1 or 100%.
112
i.e., 0 ≤ P (A) ≤ 1. This rule says that probabilities cannot be negative and as the
probability of the sample space is 1, the probability of an event contained in the sample
space should be less than or equal to 1.
Rule 3
If A and B are mutually exclusive events, then the probability of (A or B) is equal to
the sum of the probabilities of A and B.
P (A or B) = P (A) + P (B) because P (A and B) = 0 as A and B are mutually exclusive.
Mutually exclusive events are those which do not overlap when represented in Venn
diagrams (See Figure 8.1).
Figure 8.1: Mutually Exclusive Events
A B C
Two events A and B are mutually exclusive if the occurrence of one implies the non-
occurrence of the other. Hence obtaining a head on tossing a coin and obtaining a tail
are mutually exclusive events.
Figure 8.2: Rules of Probability

Probability rules
(For simultaneous trials) (For consecutive trials)

Multiplication Rule Bayes’ Theorem
Addition Rule
Mutually Partially Dependent Independent

Exclusive Overlapping events events
events events
The Addition Rule

Mutually exclusive events
This can be represented by the Venn diagram as shown in Figure 8.1.
P (A or B or C) = P (A) + P (B) + P(C)
Suppose A = getting 1 on throwing the dice
B = getting 2 on throwing the dice
C = getting 3 on throwing the dice
113
As there are six possible equally likely outcomes on throwing the dice,
P (A or B or C) = 3 = 1 + 1 + 1 = P (A) + P (B) + P(C)
6 6 6 6
Nonmutually exclusive events
If two events are not mutually exclusive the probability of one of them occurring is the
sum of the marginal probabilities of the events minus the joint probability of the
occurrence of the events.
Figure 8.3: Non Exclusive Events
C D
C and D are not mutually

exclusive events
P (A or B) = P (A) + P (B) – P (A and B)

where A and B are not mutually exclusive events.
Example 8.3
The Warwick Systems Company markets personal computers (See Table 8.1). Some
computers have two disk drives (A) and some have one disk drive (B). Another feature
of these machines is the capacity in terms of K (kilo) bytes – that is, whether they have
256K or 128K capacity. Presently, the firm’s finished goods inventory consists of 300
machines equipped with varying features (See Table 8.1). At any time, the Warwick
Systems Company may receive an order for a machine or machines with specific
features. If Warwick has a sufficient number of machines to satisfy its customers, the
customers will continue to order machines from Warwick. But if Warwick cannot
satisfy its customers’ needs, they will probably order machines elsewhere. Hence, the
management of Warwick wishes to know the likelihood that its inventories contain
machines with desirable features.
Table 8.1: Inventory of Warwick Systems Company
2DD 1DD Total
(A) (B)
With 256 K capacity (C) 100 50 150
With 128 K capacity (D) 100 50 150
Total 200 100 300
In the above example the sample space S is the set of all machines in inventory. What
is the probability of a random selection of a two-disk drive machine from inventory, or
P(A)? Also find the probability of randomly selecting a two-disk drive machine with
256K capacity.
Solution:
Let us represent two disk drive machines by A and one disk drive machine by B. We
will represent 256K by C and 128K by D.
number of ways in which A occurs 200
P (A) = = = 0.66
number of outcomes in the sample space 300
114
The probability of randomly selecting a machine with 256K capacity is:

P(C) = 150/300 = 0.5
Each of the above probabilities is designated as a marginal or unconditional
probability. Events A and C are not mutually exclusive since a machine may have both
characteristics. The probability of a machine having two disk drives or having 256K
capacity involves the addition rule with a twist. Since A and C are not mutually
exclusive events, we must apply the counting rule. Hence, the probability of A or C is:
P (A or C) = P (A) + P(C) - P (A and C) is
= 200 + 150 − 100 = 5
300 300 300 6
P (A or D) = P (A) + P(D) - P (A and D) is
= 200 + 150 − 100 = 5
300 300 300 6
Event (A or C), includes all elements except the 50 elements of B that are elements of
neither A nor C.
The probability of a machine having features B and D is:
P (B and D) = 50/300 = 0.166
The probability of the complement of (A or C) is P (B and D). These two events
account for all 300 computers.
The equation for P (A or C) sums the elements from A and C and subtracts elements
that are in both A and C.
Example 8.4
Consider a bag containing 4 white and 5 black balls. If a man draws 3 balls at random,
without replacement, what is the probability that all three are black?
Solution:
The total number of ways in which 3 balls can be drawn is 9C3and the number of ways
of drawing 3 black balls is 5C3, therefore the probability of drawing 3 black balls is
given by:
P (all three are black) = Favourable Events
Total Events
5C
= 3 = 5
9C 42
3
Example 8.5
Consider a bag containing 5 white and 7 black balls. If two balls are drawn at random
without replacement, what is the probability that one is white and the other is black?
Solution:
P (One is white and other, black)
35
= Favourable Events ( 5 C 1 )( 7 C 1 ) =
=
Total Events 12 C 66
2
Multiplication Rule
Independent events
Two events, A and B are said to be independent, if the probability of happening or not
happening of an event is not affected by the happening or not happening of the other.
115
Thus, in the case of independent events the probability of any event is not conditional,
and does not depend on the happening or not happening of the other event either
preceding, following or at the same time.
The multiplication rule of probability states that the probability of happening of an
independent event or unconditional event is given by the product or the probabilities of
each event.
Symbolically,
P (A ∩ B) = P (A). P (B)
or P (A and B) = P (A). P (B)
Thus, in the case P (A/B) = P (A) and P (B/A) =P (B)
Because the happening of A or B does not affect the happening of the other as both are
independent.
Example 8.6
In a survey involving 500 children, it was found that 200 children liked strawberry ice-
cream, 125 children liked chocolate ice-cream and 75 children liked both the ice-
creams. What is the probability of finding a child in the survey who neither likes
strawberry ice-cream nor chocolate ice-cream?
Solution:
Let S represent the event that a child likes strawberry ice-cream
200
Thus P (S) = = 0.4
500
Let C represent the event that a child likes chocolate ice-cream
125
Thus P (C) = = 0.25
500
75
P (S and C) = = 0.15
500
P (S or C) = P(S) + P(C) - P (S and C)
= 0.4 + 0.25 – 0.15
= 0.5
P (neither S nor C) = 1 - P (S or C)
= 1- 0.5
= 0.5
Example 8.7
An assembly line consists of drilling, welding and painting machine. The assembly line
works fine if all the three machines work properly. The probability of drilling machine
failing in an year is 0.20. The probability of welding machine failing in an year is 0.25.
The probability of painting machine failing in an year is 0.15. What is the probability
that the assembly line will fail before the end of the year?
Solution:
Let A, B and C represents the event of drilling, welding and painting machines failing
in a year respectively.
116
P(A) = 0.20 P( A ) = 1 − 0.2 = 0.80
P(B) = 0.25 P( B ) = 1 − 0.25 = 0.75
P(C) = 0.15 P( C ) = 1 − 0.15 = 0.85

As the events are independent, the probability of all the three machines not failing in a
year is given by
P( A ) × P( B ) × P( C ) = 0.8 × 0.75 × 0.85

= 0.51
Thus the probability of the assembly line failing before the end of the year
= 1 – 0.51
= 0.49
Example 8.8
Ramanujam Ferro Alloys Pvt. Ltd. has 500 employees, 20% are skilled and the rest are
unskilled. 30% of them are with the company for more than 10 years and the rest for
less. Of those who have been with the company for more than 10 years 40% are
skilled.
a. If an employee is selected at random, what is the probability that the employee is
unskilled and has been with the company for less than 10 years?
b. If an employee is selected at random, what is the probability that the employee is
either unskilled or has been with the company for more than 10 years?
Solution:
a. Given, probability that the employee is skilled, P (skilled) = 0.2
∴ Probability that the employee is unskilled, P (unskilled) = 1 - P(skilled)
= 0.8
Given, probability that the employee is with the company for more than 10 years,
P(more than 10 years) = 0.3
∴ Probability that the employee is with the company for less than 10 years,
P(less than 10 years) = 1 - P (more than 10 years)
= 0.7
∴ P(more than 10 years and unskilled) = 0.3 × (1-0.4)
= 0.18
Given that 80% are unskilled. Out of this 80%, 18% have worked with the company
for more than 10 years.
∴ P(less than 10 years and unskilled) = 0.8 – 0.18
= 0.62
b. P (unskilled or more than 10 years) = P (unskilled) + P (more than 10 years) –
P(unskilled and more than 10 years)
= 0.8 + 0.3 – 0.18
= 0.92
117
Example 8.9
A company wants to decide its strategy for the launch of a new product. It is trying to
estimate the level of competition that might exist in the market after the launch of the
new product. The top management of the company feels that three competitors A, B
and C might enter the market. There are 60% chances that competitor A will enter the
market, 30% chances that competitor B will enter, and 10% chances that competitor C
will enter.
a. What is the probability that no competitor will enter the market?
b. What is the probability that A and C will enter the market as competitors but not
B?
Solution:
Let A represents the event of competitor A entering the market
Α represents the event of competitor A not entering the market
B represents the event of competitor B entering the market
Β represent the event of competitor B not entering the market
C represents the event of competitor C entering the market
C represents the event of competitor C not entering the market
Given, P(A) = 0.6
P (B) = 0.3
P (C) = 0.1
P ( Α ) = 1- 0.6 = 0.4
P ( Β ) = 1-0.3 = 0.7
P ( C ) = 1- 0.1 = 0.9
a. P (A, B and C will not enter the market) = P ( Α ) × P ( Β ) × P ( C )
= 0.4 × 0.7 × 0.9
= 0.252
b. P (A and C will enter the market but not B) = P (A) × P (C) × P ( Β )
= 0.6 × 0.1 × 0.7
= 0.042
RANDOM VARIABLES
Before proceeding further, let us first understand the concept of random variables,
which enables us to understand the concept of probability distributions better.
Random variable is a variable that takes different values as a result of the outcomes of
a random experiment. A random variable is said to be continuous if it is allowed to
assume any value within a specified range and is said to be discrete if it is allowed to
take only a limited number of values, which can be listed. This can be further
explained through the following example. Suppose an unbiased pair of dice is tossed.
The possible outcome of the sum of the upper faces of the two dice can take any
integer value between 2 and 12. The outcome is said to be discrete because it can take
only a finite (or countable) number of values.
On the other hand, if the task is to determine the mean age of a sample of 1000 voters,
the possible outcome (X) can take any value in an interval(s) of numbers and is hence
118
continuous. It is a general practice to use capital letters for random variables and lower
case letters to indicate the actual value it takes. That is X = x.
Expected Value of a Random Variable

Imagine a situation of tossing a coin ten times and getting 6 heads out of the
experiment. The result is not always the same if the same experiment is repeated under
similar conditions and is bound to vary from experiment to experiment, though the coin
is totally unbiased.
Expected value is a fundamental idea in the study of probability distribution and is
obtained by multiplying each value that the variable can assume by the probability of
occurrence of that value and then summing up these products. Let us illustrate the
process of calculating the expected value with the help of Example 8.1
Example 8.10
The daily records of a dental clinic indicate that the number of patients arriving at the
clinic ranges from 30 to 45 per day. Table 8.2 illustrates the number of times each level
is reached during the past 100 days and the probability is for the same level to recur the
next day. Calculate the expected value of number of patients to arrive at the clinic.
Table 8.2: Number of Patients at Dental Clinic
Number of Patients Number of days the level Probability reaching the

was observed level
30 3 0.03
31 2 0.02
32 1 0.01
33 5 0.05
34 6 0.06
35 7 0.07
36 9 0.09
37 10 0.10
38 12 0.12
39 11 0.11
40 9 0.09
41 6 0.06
42 5 0.05
43 8 0.08
44 2 0.02
45 4 0.04
Solution:
To obtain the expected value of patients, we have to multiply each value that the
variable can assume with the probability of occurrence of that value and then sum these
products. This is illustrated in Table 8.3.
However, the expected value in the table does not mean that 38.05 patients will arrive
the next day. This only helps the dentist as a basis for his decisions on daily visits
119
because the expected value is a weighted average of the outcomes that can be expected
in the future. The dentist should recompute the expected value and update his
information on a regular basis.
Table 8.3: Calculation of Expected Value
Number of Patients (1) Probability reaching the level (1) × (2)

(2)
30 0.03 0.9
31 0.02 0.62
32 0.01 0.32
33 0.05 1.65
34 0.06 2.04
35 0.07 2.45
36 0.09 3.24
37 0.10 3.70
38 0.12 4.56
39 0.11 4.29
40 0.09 3.60
41 0.06 2.46
42 0.05 2.10
43 0.08 3.44
44 0.02 0.88
45 0.04 1.8
Expected Value of Patients = 38.05
TYPES OF PROBABILITY DISTRIBUTIONS

Probability distributions are basically of two types:
• Discrete probability
• Continuous probability
A discrete probability can take only a limited number of values, which can be listed.
The probability of taking birth in a given month is discrete because there are only 12
possible values (12 months of the year) in the distribution. Binomial and Poisson
distributions are the examples of discrete distributions. On the other hand, in a
continuous probability distribution, the variable is allowed to take on any value within
a given range. Normal distribution is an example of continuous distribution.
Discrete Probability Distributions

Since each value of a discrete random variable is linked to an outcome of an
experiment, the values of a random variable can be related to the probabilities of
outcomes. The result of this process is called a discrete probability distribution. To
illustrate the concepts of discrete probability distributions, let us consider an
experiment of tossing a balanced coin thrice. If the aim is to determine the number of
times that the head occurs (X), the results can be depicted as given in Table 8.4.
120
The values given in the relative frequency column of the Table 8.4 are nothing but the
probabilities associated with the values of X. So, the above findings can be slightly
modified as shown in Table 8.5.
Table 8.4: Results of Tossing a Balanced Coin
X Frequency Relative Frequency
0 1 1/8
1 3 3/8
2 3 3/8
3 1 1/8
Table 8.5: Probabilities of Getting Heads

X P (X = x)
0 1/8
1 3/8
2 3/8
3 1/8
Total 1
Table 8.5 is a typical example of a discrete probability distribution. One may view this
distribution as eight numbers (for instance, eight students taking a 3-subject exam in
which one failed in all, 3 got through one subject, and so on). The mean of these
numbers is calculated as below.
Mean (µ) = 0 + 1 + 1 + 1 + 2 + 2 + 2 + 3
8
µ = 1.5
The above equation can also be written as
µ = 0 × 1 + 1× 3 + 2 × 3 + 3 × 1
8 8 8 8
This can also be generalized to any discrete probability distribution as:
µ = x1P(X = x1) + x2 P(X = x2 ) + ..... + xn P(X = xn )
n
µ = ∑ x i P( x = x i )
i =1
µ = 1.5 has a practical interpretation. If this experiment were repeated an infinite

number of times and the values of X were recorded, then theoretically, 1.5 would
represent the average number of times heads would come up if the experiment is
repeated an infinite number of times. For this reason the mean is often called the
expected value E(X).
Continuous Probability Distribution

In such distributions, the variable can assume any value within a given range.
Therefore it is impossible to list all possible values. If we were studying the level of
effluents in a variety of streams, and measured the level of effluents by parts of
effluents per million parts of water. It would be expected that a continuous range of
parts per million (ppm), from very low levels upstream to extremely high levels
121
downstream. We can see therefore that the parameter, ppm can take on a very large
number of values. This distribution would therefore be called a continuous probability
distribution as all values are so close to each other that we cannot single out a group of
them.
THE NORMAL DISTRIBUTION

The normal distribution reflects the various values taken by many real life variables
like the heights and weights of people or the marks of students in a large class. In all
these cases a large number of observations are found to be clustered around the mean
value and their frequency drops sharply as we move away from the mean in either
direction. For example, if the mean height of an adult in a city is 6 feet then a large
number of adults will have heights around 6 feet. Relatively few adults will have
heights of 5 feet or 7 feet.
Further, if we draw samples of size n (where n is a fixed number over 30) from any
population, then the sample mean X will be (approximately) normally distributed with
a mean equal to µ – the mean of the population.
The characteristics of normal probability distribution with reference to the Figure 8.4
are:
• The curve has a single peak; thus it is unimodal.
• The mean of a normally distributed population lies at the center of its normal
curve.
• Because of the symmetry of the normal probability distribution, the median and
the mode of the distribution are also at the center.
• The two tails of the normal probability distribution extend indefinitely and never
touch the horizontal axis.
Figure 8.4: Frequency Curve for the Normal Probability Distribution

Mean
Median
Mode Normal probability
distribution is symmetrical
around a vertical line
erected at the mean
Left-hand tail Right-hand tail

extends indefinitely extends indefinitely
but never reaches but never reaches the
the horizontal axis horizontal axis
If σ is the standard deviation of the normal distribution, 68% of the observation will be
in the interval µ -1σ to µ +1σ.
95.5% of the observations will lie in the interval µ -2σ to µ + 2σ
99.7% of the observations will lie in the interval µ -3σ to µ + 3σ
122
The Standard Normal Distribution

The Standard Normal Distribution is a normal distribution with a mean µ = 0 and a
standard deviation σ = 1. The observation values in a standard normal distribution are
denoted by the letter Z.
Example 8.11
A population is normally distributed with mean = 0 and standard deviation = 1. What is
the probability that an observation from the population will have a value between –1.28
and 1.28?
Solution:
We know that for a normal distribution 80% of the observations lie between µ– 1.28 σ
and µ+ 1.28σ.
Here, µ= 0 and σ = 1. i.e., it’s a standard normal distribution
So 80% of the observations will lie between –1.28 σ and + 1.28 σ (from normal table)
Hence the probability that an observation will have a value between –1.28 and 1.28 is
80%.
Example 8.12
What is the probability that an observation from a standard normal distribution will lie
in the interval –1.96σ to 1.96σ ?
Solution:
From normal table, the probability is 95%.
Example 8.13
What is the probability that an observation from a standard normal distribution will lie
between –2.33σ and + 2.33σ ?
Solution:
From normal table, the probability is 98%.
Standardizing Normal Variables

Suppose we have a normal population. We can represent it by a normal variable X.
Further, we can convert any value of X into a corresponding value Z of the standard
normal variable, by using the formula
X−μ ,
Z=
σ
Where
X = the value of any random variable
µ = the mean of the distribution of the random
σ = the standard deviation of the distribution
Z = the number of standard deviations from x to the mean of the distribution
and is known as the z score or standard score.
Example 8.14
A normal variable X has a mean of 56 and a standard deviation of 12. Find the Z value
corresponding to the X value of –5.
123
Solution:
X −μ − 61 = -5.08
Z= =
σ 12
Example 8.15
A normal variable has a mean of 10 and a standard deviation of 5. What is the
probability that the normal variable will take a value in the interval 0.2 to 19.8?
Solution:
Probability (0.2 < X < 19.8)
= Probability ⎛⎜ 0.2 − 10 ≤ Z ≤ 19.8 − 10 ⎞
⎟⎟
⎜ 5 5
⎝ ⎠
= Probability (-1.96 Ρ Z Ρ 1.96)

= 95%
[Because 95% of the area under the standard normal curve lies in the interval -1.96 to
1.96]
We can see this from the Normal Table:
Area under the standard normal curve between 0 and 1.96 is 0.4750.
Due to symmetry of the standard normal distribution, area under the curve between –
1.96 and + 1.96 is twice the area under the curve between 0 and + 1.96.
Probability (–1.96 Ρ Z Ρ + 1.96) = 0.95 or 95%
Any normal variable can be converted into a standard normal variable as illustrated
above. Hence, we can use the standard normal distribution table to find the probability
that the variable will take a value within any given interval.
Example 8.16
A normally distributed random variable has a mean and standard deviation of 400 and
50 respectively. What is the probability that the random variable will take a value
between 450 and 480?
Solution:
450 − 400 480 − 400
Probability (450 < X < 480) = Probability ( <Z< )
50 50
= Probability (1ΡZΡ1.6)
Z = 1.6
Z=1
400 450 480
Example 8.17
In a class of 60 students, 10% of the students received A grade in Business Strategy
and 20% received E grade. Students who scored more than 75 marks received A grade
124
and those who scored less than 35 marks received E grade. If student marks are
normally distributed, what is the mean and standard deviation of marks?
Solution:
The value of standardized normal variable (z) for the students who scored between
mean and 75 = 1.282
75 − μ
Therefore 1.282 =
σ
1 μ + 1.282σ = 75 (i)
30% 40%
20% 10%
35 ⇐ 75
Similarly for the students who score less than 35 marks, the value of standardized
normal variable = -0.842
35 − μ
Therefore − 0.842 =
σ
1 μ − 0.842σ = 35 (ii)
Solving (i) and (ii)

We get, μ = 51 and σ = 18.67
SUMMARY
The concepts of probability were first developed to assist gamblers in games of chance
such as throwing dice or tossing a coin. However, the concepts were later applied in
several areas of management and space technology.
Probability concepts can be broadly classified into three majority types based on the
approaches to the study of probability theory: classical approach, relative frequency
approach and subjective approach. This chapter discussed the rules of probability.
Probability distributions are considered as theoretical frequency distributions. As these
distributions deal with expectations, they are used in making inferences and decisions
under uncertainty conditions.
Random variable is a variable that takes different values as a result of the outcomes of
a random experiment. The expected value of a random variable is obtained by
multiplying each value that the variable can assume with the probability of occurrence
of that value and then summing up these products. This chapter discusses the concepts
of random variables and their expected values.
Probability distributions can be broadly classified into two types: discrete and
continuous. While, binomial and Poisson distributions are discrete in nature, normal
distribution is continuous in nature.
125
Chapter 9
Matrices
• Definition of Matrix
• Types of Matrices
• Operations on Matrices
• Determinant of a Matrix
• Properties of Determinants
• Adjoint of a Matrix
• Inverse of a Matrix
• Application of Matrix Methods to Solve Linear Equations
Matrices
Matrices play an important role in modern techniques for quantitative analysis of

managerial decisions. They provide a compact way of representing a system of
equations. Matrices lay down a systematic method for presenting huge quantities of
data. Matrix algebra can also be applied to solve system of linear equations. Thus,
matrices play a vital role in a managers’ decision making process.
This chapter will introduce the concept of matrices and the types of matrices in detail.
It also discusses how various mathematical operations can be done on matrices. Later,
it introduces the concepts of determinant, adjoint and inverse of a matrix. The chapter
will also discuss how managers can use matrix methods like Cramer’s rule, matrix
inversion method and Gauss - Jordan elimination methods to solve linear equations
related to various managerial problems.
DEFINITION OF MATRIX
Matrix is a set of ‘m x n’ numbers arranged in the form of a rectangular array having

‘m’ rows and ‘n’ columns. It is called an ‘m × n’ matrix. An ‘m × n’ matrix is usually
written as
⎡a 11 , a 12 , ..... a 1n ⎤
⎢ ⎥
⎢a 21 , a 22 ,.....a 2n ⎥
A = ⎢............ ⎥
⎢ ⎥
⎢............ ⎥
⎢a , a , .....a ⎥
⎣ m1 m2 mn ⎦ m×n
In compact form it can also be written as

A = [aij]m×n where iΡm , jΡn
Here, the numbers a11, a12....etc. are called the elements of the matrix. The element aij
belongs to ith row and jth column of the matrix A.
Example for 2 × 3 matrix: ⎡1 2 3 ⎤
⎢4 5 6⎥
⎣ ⎦ 2× 3
and
⎡0 1 2 4 ⎤
Example for a 3 × 4 matrix: ⎢ 0 7 8 9 ⎥
⎢ ⎥
⎢⎣ 5 7 1 2 ⎥⎦ 3× 4
TYPES OF MATRICES
The matrices can be classified into the following types:
Row Matrix
A matrix having only a single row is called a row-matrix. The following are the
examples
A = [4 6] (Row matrix of order 1 × 2)
B = [3 -4 2] (Row matrix of order 1 × 3)
Note: A row matrix can have any number of columns but it can have only one row.
127
Column Matrix
A matrix having only a single column is called a column matrix. The following are the
examples of column matrix:
⎡1 ⎤
X = ⎢3 ⎥ (Column matrix of order 3×1)
⎢ ⎥
⎢⎣ 6 ⎥⎦
3×1
⎡8 ⎤
⎢ ⎥
Y = ⎢6 ⎥ (Column matrix of order 4×1)
⎢4 ⎥
⎢ ⎥
⎣1 ⎦ 4×1
Note: A column matrix can have any number of rows but it can have only one column.
Zero or Null Matrix

A matrix in which all the elements are zeros is called a Zero or a Null matrix.
For example:
⎡0 0 0⎤
L = [0]1x1 M = ⎡⎢ 0 0 ⎤⎥ N = ⎢0 0 0⎥
0 0 ⎣ ⎦2×2
⎢ ⎥
⎢⎣ 0 0 0 ⎥⎦
3× 3
Since all the elements of the matrices L, M and N are zero, L, M and N are Null
matrices.
Square Matrix
An m × n matrix for which m = n, i.e., (the number of rows = number of columns) is
called a square matrix of order n (since m = n) or m rowed matrix. The following is a
square matrix of order 3×3.
⎡2 3 4⎤
A = ⎢5 6 7 ⎥ is a square matrix of order 3
⎢ ⎥
⎢⎣1 0 2 ⎥⎦
3× 3
If A = [aij]m x n is a square matrix of order m, then the elements aij for which i = j, are
called the diagonal elements of A (i.e. a11, a22, a33.......ann are diagonal elements). In the
above square matrix A, 2, 6, 2 are the diagonal elements.
The line along which these diagonal elements lie is called the principal diagonal or
the main-diagonal of the matrix.
Diagonal Matrix
A square matrix, in which all the elements except the diagonal elements are zero, is
called a diagonal matrix.
If A = [aij] is a diagonal matrix of order m, where aij = 0 for i ≠ j. It can be expressed
as
A = diag. [a11, a22, a33 .........amm]
⎡7 0 0⎤
Example: A = ⎢ 0 6 0 ⎥ is a diagonal matrix as all the elements other than the
⎢ ⎥
⎣⎢ 0 0 9 ⎥⎦ 3× 3
diagonal elements are zeros. It is denoted as A = diag. [7, 6, 9]
128
Matrices
Upper Triangular Matrix

A square matrix A = [aij] is called an upper triangular matrix if aij = 0, whenever i > j.
Thus in an upper triangular matrix, all the elements below the principle diagonal are
zeros. For example
⎡1 2 3 ⎤
A = ⎢0 4 1 ⎥
⎢ ⎥
⎢⎣ 0 0 2 ⎥⎦
3× 3
A triangular matrix A = [aij] is called strictly upper triangular if there are zeros along
the diagonal as well as in the lower portion. i.e. aij = 0 for i Σj
Lower Triangular Matrix

A square matrix A = [aij] is called a lower triangular matrix if aij = 0 whenever i < j.
Thus in a lower triangular matrix all the elements above the principal diagonal are
zero.
⎡2 0 0⎤
B= ⎢1 3 0 ⎥
⎢ ⎥
⎢⎣ 4 5 1 ⎥⎦
3× 3
A triangular matrix A = [aij] is called strictly lower triangular if there are zeros along
the diagonal as well as in the upper portion. i.e. aij = 0 for i Ρj
Scalar Matrix
A diagonal matrix whose diagonal elements are all equal is called as scalar matrix.
⎡2 0 0⎤
A = ⎢0 2 0⎥
⎢ ⎥
⎢⎣ 0 0 2 ⎥⎦
3× 3
Unit Matrix or Identity Matrix

A square matrix each of whose diagonal elements equal to “1” (unity) and each of
whose non-diagonal elements equal to zero is called a unit matrix or an identity
matrix and is denoted by “I”.
In denotes a unit matrix of order n.
In other words, a square matrix [aij]m×n is said to be a unit matrix if aij = 1 when i = j
and aij =0 and i ≠ j
For example
I2 = ⎡⎢1 0 ⎤⎥
⎣ 0 1 ⎦ 2× 2
⎡1 0 0 0 ⎤
⎢ ⎥
I4 = ⎢ 0 1 0 0 ⎥
⎢0 0 1 0 ⎥
⎢ ⎥
⎣ 0 0 0 1 ⎦ 4× 4
Singular and Non-Singular Matrices

A square matrix A is said to be a singular matrix when A = 0 and it is known as
non-singular when A ≠ 0.
129
Calculation of A is explained later.

For example:
21
A = ⎡⎢ 2 1 ⎤⎥ is singular as A = =0
⎣6 3 ⎦ 63
3 4
and B = ⎡3 4 ⎤ is non-singular as B = = -5 ≠ 0
⎢2 1 ⎥ 21
⎣ ⎦
OPERATIONS ON MATRICES
To use matrices, one should be conversant with mathematical operations like addition,
subtraction and multiplication on matrices. The following are some of the
mathematical operations on matrices and their properties.
Addition of Matrices
Let A and B be two matrices of the same order m×n then their sum (A+B) is defined
to be the matrix of the order m × n obtained by adding the corresponding elements of
A and B.
If A = [aij]m×n and B = [bij]m×n
Then A + B = [aij + bij]m×n
Example 9.1
If A = ⎡⎢ 2 1 ⎤⎥ and B = ⎡⎢ 3 4 ⎤⎥ Find A + B
0 1
⎣ ⎦ 1 2 ⎣ ⎦
Solution:
A + B = ⎡⎢ 2 + 3 1 + 4 ⎤⎥ = ⎡⎢ 5 5 ⎤⎥
⎣0 + 1 1 + 2 ⎦ ⎣1 3 ⎦
If A = ⎡⎢1 3 ⎤⎥ and B = ⎡⎢ 3 1 0 ⎤⎥
⎣2 4⎦ ⎣6 2 5⎦
then A + B is not defined, since A and B are not of the same order.
Properties of matrix addition
Some of the properties of matrix addition are:
Commutative property
If A and B are two matrices of same order m×n then,
A+B=B+A
Example 9.2
If A = ⎡⎢ 2 1 ⎤⎥ and B = ⎡⎢ 3 4 ⎤⎥
⎣ 0 1⎦ ⎣1 2 ⎦
Solution:
Then A + B = ⎡ 2 1⎤ + ⎡3 4 ⎤ = ⎡ 2 + 3 1 + 4 ⎤ = ⎡5 5 ⎤
⎢0 1 ⎥⎦ ⎢1 2 ⎥ ⎢0 + 1 1 + 2 ⎥ ⎢1 3 ⎥
⎣ ⎦
⎣ ⎣ ⎦ ⎣ ⎦
130
Matrices
B + A = ⎡⎢ 3 4 ⎤⎥ + ⎡⎢ 2 1 ⎤⎥ = ⎡⎢ 3 + 2 4 + 1 ⎤⎥ = ⎡⎢ 5 5 ⎤⎥
⎣1 2 ⎦ ⎣ 0 1 ⎦ ⎣1 + 0 2 + 1 ⎦ ⎣1 3 ⎦
∴A + B = B + A
Associative property
If A, B, C are three matrices of order m × n then,
(A+B) + C = A + (B + C)
Example 9.3
If A = ⎡ 2 1 ⎤ B = ⎡3 4 ⎤ C = ⎡1 2 ⎤
⎢0 1⎥ ⎢1 2 ⎥ ⎢3 4 ⎥
⎣ ⎦ ⎣ ⎦ ⎣ ⎦
Solution:
A + B = ⎡ 2 1 ⎤ + ⎡ 3 4 ⎤ = ⎡⎢ 5 5 ⎤⎥
⎢ 0 1 ⎥ ⎢1 2 ⎥ ⎣1 3 ⎦
⎣ ⎦ ⎣ ⎦
Then (A + B) + C = ⎡⎢ 5 5 ⎤⎥ + ⎡ 1 2 ⎤ = ⎡ 6 7 ⎤
⎣1 3 ⎦ ⎢3 4 ⎥ ⎢4 7⎥
⎣ ⎦ ⎣ ⎦
B + C = ⎡ 3 4 ⎤ + ⎡1 2 ⎤ = ⎡ 4 6 ⎤
⎢1 2 ⎥ ⎢3 4 ⎥ ⎢4 6⎥
⎣ ⎦ ⎣ ⎦ ⎣ ⎦
A + (B + C) = ⎡ 2 1 ⎤ + ⎡ 4 6 ⎤ = ⎡ 6 7 ⎤
⎢0 1⎥ ⎢4 6⎥ ⎢4 7⎥
⎣ ⎦ ⎣ ⎦ ⎣ ⎦
Hence (A + B) + C = A + (B + C)
Existence of additive inverse
For a matrix A = [aij]m×n its additive inverse is defined as the matrix [-aij]m×n and is
denoted by -A. Therefore, we have A + (-A) = 0 = (-A) + A
Example 9.4
⎡1 4 7 ⎤
Given the matrix M= ⎢⎢2 5 8⎥⎥ . Find the additive inverse of matrix M.
⎢⎣3 6 9⎥⎦
Solution:
For a matrix A = [aij]m×n its additive inverse is defined as the matrix [-aij]m×n and is
denoted by -A.
⎡ − 1 − 4 − 7⎤
⎢− 2 − 5 − 8 ⎥
⎢ ⎥
⎢⎣ − 3 − 6 − 9 ⎥⎦
Multiplication of a Matrix by a Scalar

Let A be a matrix and k is any scalar number. Then the m×n matrix obtained by
multiplying every element of matrix A by k is called the scalar multiple of A by k.
It is denoted by k.A or A.k
If A = [aij]m×n
131
Then k.A = A.k = [k aij]m×n

Properties of multiplication
• If A and B are two matrices each of the order m × n, then
k(A+B) = kA +kB
• If p and q are two scalars and A is any matrix of order m × n , then (p+q).A = p.A
+ q.A
• If p and q are two scalars and A is a matrix of order m × n, then p. (q.A) = (p.q).
A
• If A be any matrix of order m × n and k is a scalar, then (-k)A = - (k.A) = k.(-A)
Example 9.5
Given a matrix C = [3 9 5] and a scalar number k = 4, find kC.
Solution:
Let A be a matrix and k be a scalar number. Then, the matrix obtained by multiplying
every element of matrix A by k is called the scalar multiple of A by k. It is denoted by
kA or Ak
14C = [12 36 20]
Multiplication of Matrices
Let A = [aij]m×n and B = [bjk]n×p
be two matrices such that the number of columns in A is equal to the number of rows
in B. Then the product of the two matrices is given by
C = [cik ]m× p
m n p
where cik = ∑ ∑ ∑ a ij b jk (product of A and B in that order)
i =1 j=1k =1
and is denoted as C = AB
Thus, for two matrices A and B, if A is an m × n matrix and B is an n × p matrix, then
AB will be an m × p matrix.
The (ik)th element of AB is obtained by multiplying each element of ith row of A by
the corresponding element in the kth column of B and then adding the products.
Example 9.6
⎡b11 b12 ⎤
⎡a 11 a 12 a 13 ⎤ ⎢ ⎥
If A = ⎢ ⎥ and B = ⎢b 21 b 22 ⎥
⎣a 21 a 22 a 23 ⎦ ⎢⎣b 31 b 32 ⎥⎦
⎡b11 b12 ⎤
⎡a 11 a 12 a 13 ⎤ ⎢ ⎥
Then AB = ⎢ ⎥ ⎢b 21 b 22 ⎥
⎣a 21 a 22 a 23 ⎦ ⎢⎣b 31 b 32 ⎥⎦
⎡c11 c12 ⎤
Let, AB = ⎢ ⎥
⎣c 21 c 22 ⎦
where c11 = (1st row of A) × (1st column of B)
= a11b11 +a12 b21 +a13b31
c12 = (1st row of A) × (2nd column of B)
132
Matrices
= a11b12 +a12 b22 +a13b32

c21 = (2nd row of A) × (1st column of B)
= a21b11 +a22 b21 +a23b31
c22 = (2nd row of A) × (2nd column of B)
= a21 b12 + a22 b22 +a23 b32
⎡a 11b11 + a 12 b 21 +a 13 b 31 a 11b12 + a 12 b 22 +a 13 b 32 ⎤
AB= ⎢ ⎥
⎣a 21b11 + a 22 b 21 +a 23 b 31 a 21b12 + a 22 b 22 +a 23 b 32 ⎦
Note: A product of two matrices A and B (i.e. AB) is defined only when the number
of columns in matrix A is equal to the number of rows in matrix B.
Example 9.7
⎡2 3⎤ ⎡4 5⎤
Given P = ⎢ ⎥ and Q = ⎢2 3⎥ . Find PQ.
⎣ 5 1⎦ 2×2 ⎣ ⎦ 2×2
Solution:
⎡2 3⎤ ⎡4 5⎤
PQ = ⎢ ⎥×⎢ ⎥
⎣5 1⎦ ⎣2 3⎦
⎡(2 × 4) + (3 × 2) (2 × 5) + (3 × 3)⎤
= ⎢ ⎥
⎣ (5 × 4) + (1 × 2) (5 × 5) + (1 × 3) ⎦
⎡ 8 + 6 10 + 9 ⎤
= ⎢ ⎥
⎣20 + 2 25 + 3⎦
⎡14 19 ⎤
= ⎢ ⎥
⎣22 28⎦
Properties of matrix multiplication
• Matrix multiplication is associative i.e., when A,B,C are three matrices of order
m × n, n × p, p × q respectively, A.(B.C) = (A.B).C
• Matrix multiplication is distributive with respect to operation “+”, i.e. when A, B,
C are three matrices of the order m × n, n × p, n × q then,
A. (B+C) = A.B+A.C
• The matrix multiplication is not always commutative. i.e. AB ≠ BA.
DETERMINANT OF A MATRIX
Determinant is a value of a matrix and is denoted by
det A or A , and is written as
a11,a12,.....a 1n
a 21,a 22,.....a 2 n
det A = ............
............
a n1,a n 2,.....a nn
133
det of A is also represented by the symbol Δ

Note: The determinant can be defined only for square matrices.
Value of Determinant of Order 2

a11 a12
A = Δ = is the determinant of order 2
a 21 a 22
The value of the determinant of order 2 is calculated as a11.a22 - a12.a21
In other words the value of the determinant of order 2 is equal to the product of the
(elements along the principle diagonal) - (The product of the off diagonal elements).
Example 9.8
⎡4 2⎤
[A] = ⎢ ⎥
⎣1 3⎦
Solution:
A = 4(3) - 1 (2) = 12 - 2 = 10
i.e. A = 10
Minors of an Element
For a determinant, the minor Mij of an element aij can be defined as the value of the
determinant obtained by leaving the row and the column passing through the element
aij (i.e. deleting the ith row and jth column of the determinant). The second order
determinant thus obtained (in case of 3×3 matrix) is called the minor of that element
aij and it is denoted by Mij.
a 1 b1 c1
For example, consider Δ = a 2 b 2 c 2
a 3 b3 c3
Then the minor of the element a1 is calculated by deleting the row and column in
which it is present (i.e. first row and first column of the above determinant). Then the
minor M11 of a1 is given by
b2 c2
M11 = = (b2 c3 - b3 c2)
b3 c3
Similarly the minors of other elements b1, c1.....c3 can be calculated.

Example 9.9
⎡ 1 2 3⎤
Given a matrix A = ⎢⎢4 5 6⎥⎥ . Calculate the minor for the element ‘6’.
⎢⎣7 8 9⎥⎦
Solution:
The minor of an element is calculated by deleting the row and column in which it is
present.
1 2
∴ M23 = = 8-14 = -6
7 8
134
Matrices
Co-factor of an Element
The minor Mij multiplied by (-1)i+j is called the co-factor of the element aij.
It is denoted as
Co-factor of aij = Aij = (-1)i+j . Mij
a11 a12 a13
If Δ = a 21 a 22 a 23
a 31 a 32 a 33
The co-factor of
a 22 a 23
a11 = A11 = (-1)1+1 = M11
a 32 a 33
a 21 a 23
a12 = A12 = (-1)1+2 = –M12
a 31 a 33
Similarly, co-factors of other elements can be calculated.

Example 9.10
⎡1 2 3 ⎤
If A = ⎢⎢4 5 6⎥⎥ , find out the co-factors of elements 1,2 and 3.
⎣⎢7 8 9 ⎦⎥ 3×3
Solution:
56
Co-factor of element 1 = (-1)1+1 = 45 - 48 = -3
89
46
Co - factor of element 2 = (-1)1+2 = -1(36 - 42) = 6
79
45
Co - factor of element 3 = (-1)1+3 = 32 - 35 = -3
78
NOTE: For a given determinant the sum of the products of the elements of any row or
column with their corresponding co-factors gives the Value of the determinant.
Value of Determinant of Order 3

a 11 a 12 a 13
Δ = a 21 a 22 a 23
a 31 a 32 a 33
The value of this determinant of order 3 can be found by expanding the determinant
along the 1st row.
a 22 a 23 a 21 a 23 a 21 a 22
= a11 - a12 + a13
a 32 a33 a 31 a 33 a31 a32
135
i.e., Δ = a11A11 + a12 A12 +a13 A13

Where A11, A12, A13 are the co-factors of a11, a12 and a13 respectively.
Determinant of Order n:
A determinant of order ‘n’ has n rows and n columns i.e. n×n elements. A determinant
of order n is a square array of n×n numbers enclosed between vertical bars.
a 11 , a 12 ,.....a 1n
a 21 , a 22 ,.....a 2 n
Δ= ............
............
a n1 , a n 2 ,.....a nn
The co-factor Aij of element aij in Δ

= (-1)i+j times the determinant of order (n-1) obtained from Δ by leaving the row and
the column passing through the element aij.
∴ Δ = ai1 Ai1 +ai2 Ai2.............+ainAin [where i=1 to n]
Δ = a1j A1j +a2j A2j.............. +anjAnj [where j=1 to n]
PROPERTIES OF DETERMINANTS
Some of the properties of determinants are as follows:

• The value of a determinant does not change when rows and columns are
interchanged.
a1 b1 c1 a1 a 2 a 3
i.e, a 2 b 2 c 2 = b1 b 2 b 3
a 3 b3 c3 c1 c 2 c 3
• When any 2 rows or columns of determinant are interchanged, the value of the
determinant is multiplied by (-1).
C1 ⇔ C3
a1 b1 c1 c1 b1 a1
i.e., a 2 b 2 c 2 = (-1) c 2 b 2 a 2
a 3 b3 c3 c3 b 3 a 3
• When all the elements of one row or one column of a determinant are multiplied
by the same number ‘k’ the value of the new determinant is k times the value of
the given determinant.
ka1 b1 c1 a1 b1 c1
i.e., ka 2 b 2 c 2 = k a 2 b 2 c 2
ka 3 b 3 c 3 a 3 b3 c3
In the above determinant all the elements of the 1st column are multiplied by k.
• If two rows or two columns of a determinant are identical then the value of the
determinant is zero.
136
Matrices
a1 b1 a1
i.e. a 2 b 2 a 2 = 0
a 3 b3 a 3
Since column one is identical to column three.
• In a determinant the sum of the products of the elements of any row or column
with the co-factors of any other row or column elements is zero.
a 1 b1 c1
Let Δ = a 2 b2 c2
a 3 b3 c3
Let A1, B1, C1, A2, B2, C2, A3, B3 and C3, are the co-factors of the elements a1, b1, c1, a2
in Δ .
...c3
Then Δ = a1A1 +b1 B1 +c1C1

But when we multiply a1, b1, c1, with co-factors of any other row the product thus
obtained will be zero
i.e., a1A2 +b1 B2 +c1C2 = 0
• In a determinant when each element of a row or a column is a sum of two terms,
then the determinant can be expressed as the sum of two determinants of the same
order.
a 1 + α1 b1 c1
In a 2 + α1 b 2 c 2
a 3 + α1 b 3 c 3
The elements of column one are all sum of two terms. So it can be expressed as sum
of two determinants of same order.
a1 b1 c1 α1 b1 c1
a 2 b 2 c 2 + α1 b 2 c 2
a 3 b3 c3 α1 b 3 c 3
• When the elements of a row or column are added with ‘m’ times the
corresponding elements of another row or column then the value of the
determinant thus obtained is equal to the value of the original determinant.
a 1 b1 c1
i.e., If Δ = a 2 b2 c2
a 3 b3 c3
a 1 + mb1 b1 c1 a 1 b1 c1
Then a 2 + mb 2 b 2 c 2 = a 2 b 2 c 2
a 3 + mb 3 b 3 c 3 a 3 b3 c3
ADJOINT OF A MATRIX
When [aij]n×n is a square matrix of order n and Aij is the co-factor of aij. Then, the
adjoint of A, denoted by adj A is defined as adj A= [Aji]n×n, i.e., adj A is the transpose
of the matrix of corresponding co-factors of elements of det.A
137
T
⎡A11 A12 A13 ⎤ ⎡A11 A 21 A 31 ⎤
⎢ ⎥ ⎢ ⎥
Then adj A = ⎢A 21 A 22 A 23 ⎥ = ⎢A12 A 22 A 32 ⎥
⎢⎣A 31 A 32 A 33 ⎥⎦ ⎢⎣A13 A 23 A 33 ⎥⎦
A11, A12...........A33 are the co-factors of the corresponding elements a11, a12...........a33
Example 9.11
⎡a 11 a 12 a 13 ⎤ ⎡1 1 1 ⎤
⎢ ⎥
If A = ⎢a 21 a 22 a 23 ⎥ = ⎢⎢1 2 − 3⎥⎥ , find the adjoint of the matrix A.
⎢⎣a 31 a 32 a 33 ⎥⎦ ⎢⎣2 − 1 3 ⎥⎦
Solution:
In the above example
a11 = 1, a12 = 1, a13 = 1, a21 = 1, a22 = 2...........a33 = 3
And co-factors are A11 = 3, A12 = 9, A13 = –5...A33 = 1
T
⎡ 3 − 9 − 5⎤ ⎡ 3 − 4 − 5⎤
Hence, adj A = ⎢⎢− 4 1 3 ⎥⎥ = ⎢⎢− 9 1 − 4⎥⎥
⎢⎣ − 5 − 4 1 ⎥⎦ ⎢⎣ − 5 3 1 ⎥⎦
INVERSE OF A MATRIX
When A is any n-rowed, non-zero square matrix, then a matrix B exists such that
A.B = B.A = In
Then, B is called the inverse of matrix A
Formula for finding A-1
1
A-1 = (adj A)
A
Example 9.12
⎡0 1 2 ⎤
For A = ⎢⎢1 2 3⎥⎥ Find A-1
⎢⎣3 1 1 ⎥⎦
A = 0 – 1(1–9) + 2 (1–6) = 8 – 10 = –2
∴ A ≠0
Since A ≠ 0, A inverse exists (A-1 exists)
Finding adj.A,
A11 = Co-factor of ‘0’ = (–1)1+1 2 3 = 1(2–3) = – 1

1 1
A12 = Co-factor of ‘1’ = (–1)1+2 1 3 = – 1(1– 9) = 8

3 1
138
Matrices
A13 = Co-factor of ‘2’ = (–1)1+3 1 2 = 1(1-6) = -5

3 1
A21 = Co-factor of ‘1’ = (– 1)2+1 1 2 = –1(1–2) = +1

1 1
A22 = Co-factor of ‘2’ = (– 1)2+2 0 2 = 1(0– 6) = – 6

3 1
A23 = Co-factor of ‘3’ = (-1)2+3 0 1 = – 1(0–3) = +3

3 1
A31 = Co-factor of ‘3’ = (-1)3+1 1 2 = 1(3– 4) = – 1

2 3
A32 = Co-factor of ‘1’ =(-1)3+2 0 2 = –1(0– 2) =2

1 3
A33 = Co-factor of ‘1’ = (-1)3+3 0 1 = +1(0–1) = – 1

3 1
T
⎡A11 A12 A13 ⎤
⎢ ⎥
Then Adj A = ⎢A 21 A 22 A 23 ⎥
⎢ ⎥
⎣⎢A 31 A 32 A 33 ⎦⎥
T
⎡ − 1 8 − 5⎤ ⎡ − 1 1 − 1⎤
= ⎢⎢ 1 − 6 3 ⎥⎥ = ⎢⎢ 8 − 6 2 ⎥⎥
⎢⎣ − 1 2 − 1 ⎥⎦ ⎢⎣ − 5 3 − 1⎥⎦
1
A-1 = (adj A)
A
⎡ − 1 1 − 1⎤
−1 ⎢
= 8 − 6 2 ⎥⎥
2 ⎢
⎢⎣ − 5 3 − 1⎥⎦
⎡1 1 1 ⎤
⎢2 − 2 2 ⎥
⎢ ⎥
= ⎢− 4 3 − 1 ⎥
⎢5 3 1⎥
⎢ − ⎥
⎣2 2 2⎦
Rank of a Matrix
The number ‘r’ is said to be a rank of a matrix A if it possesses the following two
properties.
1. There is at least one sub matrix of A of order ‘r’ whose determinant is not equal
to zero.
2. If the matrix A contains any square sub matrix of order r+1 then the determinant
of every square sub matrix of order r+1 should be zero (i.e. the rank of a matrix is
the order of any highest order non vanishing minor of the matrix.)
139
The rank of a matrix A is denoted by the symbol ρ (A)

For example,
⎡1 0 0 ⎤
A = ⎢⎢ 0 1 0 ⎥⎥
⎢⎣ 0 0 1⎥⎦
3×3
We have A ≠ 0
∴ A is a non singular matrix, hence rank of A,
i.e., ρ (A) = 3
Note: The rank of every non singular matrix of order ‘n’ is ‘n’ itself.
APPLICATION OF MATRIX METHOD TO SOLVE LINEAR EQUATIONS
Linear equations can be solved by using matrix methods like: matrix inversion
method, Cramer’s Rule and Gauss - Jordan Elimination methods. However, before
looking into these methods, one should be conversant with how to present the linear
equations in matrix notation.
Presenting Linear Equations in Matrix Notation

Consider the following n linear equations in n unknowns:
a11x1 +a12x2+..............+ a1n xn = b1
a21x1+a22x2+...............+ a2nxn = b2
.....................................................
.....................................................
an1x1+an2x2+...............+annxn = bn
The above system of linear equations can be presented in matrix for A X = B
a11 , a12 ,..... a1 n
a 21 , a 22 ,..... a 2 n
Where, A = ............
............
a n 1 , a n 2 ,..... a nn
x1
x2
X = ..... and,
.....
xn
b1
b2
B = ...
...
bn
140
Matrices
Cramer’s Rule
As described earlier, determinants are useful in finding the inverse of a non-singular
matrix. Now, let us see its use in solving linear systems whose matrix coefficient (A)
is non-singular (or invertible).
For example, consider a linear system in matrix form,
A X = B (with two unknowns)
Where, A is the matrix coefficient
B is the non-homogeneous term and
X, the unknown column-matrix
Consider the system,
a1x +b1y = c1............. Equation (i)
a2x +b2y = c2.............. Equation (ii)
Multiplying (i) by b2 and (ii) by b1and subtracting,
We get (a1b2 - a2b1) x = (b2c1 – b1c2)
(b 2 c1 − b1c 2 )
∴x =
(a 1b 2 − a 2 b1 )
c1 b1
c 2 b2
=
a 1 b1
a 2 b2
Now multiply equation (i) by a2 and equation (ii) by a1 and subtracting thus obtained
equations we get (a2b1-a1b2)y = (a2c1 - a1c2)
c1 a 1
(a 2 c1 − a 1c 2 ) c2 a 2
∴y = = −
(a 2 b1 − a1b 2 ) a 1 b1
a 2 b2
a1 b1 c1 b1 c1 a1
Let D = ; D1 = ; D2 =
a 2 b2 c 2 b2 c2 a2
D1 D2
Then, x = and y= −
D D
This is called as Cramer’s rule.
The values of D, D1, and D2 can also be obtained as follows:
a1 b1
D = det.A = i.e. det of coefficients of x and y
a 2 b2
⎛ a1 ⎞
D1 is obtained from A by replacing the first column ⎜ ⎟
⎜a ⎟
⎝ 2⎠
141
⎛ c1 ⎞
by column B i.e., ⎜ ⎟
⎜c ⎟
⎝ 2⎠
c1 b1
we get D1 =
c 2 b2
D2 is obtained from A by replacing the second column
⎛ b1 ⎞ ⎛ c1 ⎞
i.e. ⎜ ⎟ ⎜ ⎟ and taking –ve of it.
⎜ b ⎟ by column B i.e. ⎜c ⎟
⎝ 2⎠ ⎝ 2⎠
a 1 c1
we get D2 =
a2 c2
Example 9.13
Solve the linear system using Cramer’s rule,
7x -5y = 20
2x – y = 7
Solution:
Writing the above system in matrix form A X = B
⎡7 − 5⎤ ⎡ x ⎤ ⎡20⎤
⎢ 2 − 1 ⎥ ⎢ ⎥ = ⎢7 ⎥
⎣ ⎦ ⎣ y⎦ ⎣ ⎦
Using Cramer’s rule, we have.
⎡ 7 − 5⎤
D= ⎢ ⎥ = - 7 - (-10) = 3
⎣2 − 1 ⎦
⎡20 − 5⎤
D1 = ⎢ ⎥ = - 20 – (- 35) = 15
⎣7 − 1 ⎦
⎡20 7 ⎤
D2 = ⎢ ⎥ = 40 – 49 = -9
⎣7 2 ⎦
D1 15
x= = =5
D 3
D 9
y= − 2 = =3
D 3
Cramer’s Rule: (Three unknowns or 3×3 matrix)
Let the system of equations be
a1X +b1Y+c1Z =d1
a2X +b2Y+c2Z =d2
a3X +b3Y+c3Z =d3
⎡ a1 b1 c1 ⎤ ⎡ d1 ⎤ ⎡X ⎤
⎢ ⎥ ⎢ ⎥
A = ⎢ a 2 b2 c 2 ⎥ B = ⎢ d 2 ⎥ X= ⎢⎢Y ⎥
⎥
⎢a b c ⎥ ⎢d ⎥ ⎢⎣ Z ⎥⎦
⎣ 3 3 3⎦ ⎣ 3 ⎦
142
Matrices
a1 b1 c1
Where D = a 2 b2 c 2 ≠0
a 3 b3 c 3
d1 b1 c1
Then D1 = d 2 b 2 c 2
d 3 b3 c 3
Obtained by replacing first column by B

a1 d1 c1
D2 = a 2 d 2 c 2
a3 d 3 c3
Obtained by replacing second column by B

a1 b1 d1
D3 = a 2 b2 d 2
a 3 b3 d 3
Obtained by replacing third column by B

D1 D D3
X= ,Y= 2 Z=
D D D
Example 9.14
5x – 2y + 3z = - 1
3x + y – 2z = 25
2x - 4y + 5z = - 29
Solution:
⎡5 − 2 3 ⎤
D= ⎢3 1 − 2⎥
⎢ ⎥
⎢⎣2 − 4 5 ⎥⎦
1 −2 3 −2 3 1
=5 – (–2) +3
−4 5 2 5 2 −4
= 5(5 – 8) + 2 (15– (– 4)) + 3(-12 – 2)
= 5(– 3) + 2(19) + 3 (–14)
= –15 + 38 – 42 = –19
D = –19
−1 −2 3
D1 =
25 1 −2
− 29 − 4 5
−2 3
= –1 1 −2 – 25 – 29 − 2 3
−4 5 −4 5 1 −2
143
= –1(5 – 8) –25 (–10 –(– 12)) –29 (4–3)

=-1(–3) –25(2) –29
= 3 –50 –29
D1= –76
5 −1 3
D2 = 3 25 − 2
2 − 29 5
3 −2 5 3 5 3
= – (–1) + 25 –(–29)
2 5 2 5 3 −2
= 1(15 – (–4)) + 25 (25–6) +29 (–10 –9)

= 19 + 25 (19) + 29 (–19)
=19(1 + 25 – 29) = 19(–3)
D2 = –57
5 − 2 −1
D3 = 3 1 25
2 − 4 − 29
3 1 5 −2 5 −2
= –1 – 25 – 29
2 −4 2 −4 3 1
= –1 (–12 –2) –25 (–20 – (–4)) –29 (5 – (–6))

= –1(–14) –25 (–16) –29 (11)
=14 + 400 – 319 = 95
D3 = 95
D = –19, D1 = –76, D2 = –57, D3 = 95
D1
∴X = =4
D
D2
Y= =3
D
D3
Z= = -5
D
Matrix Inversion Method
Matrix inversion method can be used to solve system of equations. The following
example illustrates how one can solve a system of equations using this method.
Example 9.15
Solve
x+y+z=6
x–y+z=2
2x – y + 3z = 9, using Matrix inversion method.
144
Matrices
Solution:
The given system of equations can be expressed as,
AX = B
⎡1 1 1 ⎤ ⎡x⎤ ⎡6 ⎤
Where A = ⎢1 − 1 1 ⎥ , X = ⎢ y ⎥ , B = ⎢2⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢⎣ 2 − 1 3 ⎥⎦ ⎢⎣ z ⎥⎦ ⎢⎣ 9 ⎥⎦
Now, A = – 2 ≠ 0
∴ A-1 exists and
A-1 = 1 (adj A)
A
⎡− 2 − 4 2 ⎤
= −1 ⎢ −1 1 0 ⎥⎥
2 ⎢
⎢⎣ 1 3 − 2 ⎥⎦
Then AX = B
⎡− 2 − 4 2 ⎤ ⎡6⎤
⇒ X = A B = − 1 ⎢⎢ − 1
-1
1 0 ⎥⎥ ⎢2⎥
⎢ ⎥
2 ⎢
⎣ 1 3 − 2 ⎥⎦ ⎢⎣ 9 ⎥⎦
⎡x ⎤
⎡− 2⎤ ⎡1 ⎤
⇒ ⎢⎢ y ⎥⎥ = − 1 ⎢− 4 ⎥⎥ = ⎢⎢ 2 ⎥⎥
2 ⎢
⎢⎣ z ⎥⎦ ⎢⎣ − 6 ⎥⎦ ⎢⎣ 3 ⎥⎦
⇒ x = 1, y = 2, and z = 3
Gauss - Jordan Elimination Method
The procedure for solving linear equations using Gauss- Jordan Elimination method is
as follows:
1. Write the system of equations as an augmented matrix.
2. Identify the left most column that is not all zeros.
3. If the top element in the left most column is zero interchange the top row to bring
a non-zero element in that column.
4. If that non-zero element is ‘a’, then multiply top row by 1/a to get a leading 1 in
that row.
5. Add multiples of this row to other rows so that all other rows have a 0 in this
column.
6. Cover (ignoring) the top row and go back to the second step, considering the rows
below this one (until step 5).
Continue this process until the matrix is reduced to a Unit matrix.
The following example illustrates the procedure to solve system of equations using
Gauss - Jordan Elimination Method.
Example 9.16
Find the values of x, y, and z by solving the following system of equations.
145
2y - 3z = 2
2x + z = 3
x - y + 3z = 1
Solution:
Step 1: Write the system of equations as an augmented matrix
⎡0 2 − 3 2⎤
⎢2 0 1 3⎥⎥
⎢
⎢⎣1 − 1 3 1 ⎥⎦
Step 2: Interchange the first and third row to make the top left element non- zero.
⎡1 − 1 3 1 ⎤
⎢2 0 1 3⎥⎥
⎢
⎢⎣0 2 − 3 2⎥⎦
Step 3: Add -2 times of first row to second row to get 0 at the second element of the
first column.
⎡1 − 1 3 1⎤
⎢0 2 − 5 1 ⎥
⎢ ⎥
⎢⎣0 2 − 3 2⎥⎦
Step 4: Divide second row by 2 to get a leading 1 in row 2

⎡1 − 1 3 1 ⎤
⎢0 1 − 5 / 2 1 / 2 ⎥
⎢ ⎥
⎢⎣0 2 −3 2 ⎥⎦
Step 5: Add second row to first; and add -2 times second row to third to get 0’s in the
second column
⎡1 0 1 / 2 3 / 2⎤
⎢0 1 − 5 / 2 1 / 2 ⎥
⎢ ⎥
⎢⎣0 0 2 1 ⎥⎦
Step 6: Divide row 3 by 2 to get a leading 1
⎡1 0 1 / 2 3 / 2⎤
⎢0 1 − 5 / 2 1 / 2 ⎥
⎢ ⎥
⎢⎣0 0 1 1 / 2 ⎥⎦
Step 7: Add -1/2 times of row 3 to row 1 and add 5/2 times row 3 to row 2 to get 0’s
in the third column.
⎡1 0 0 5 / 4 ⎤
⎢0 1 0 7 / 4 ⎥
⎢ ⎥
⎢⎣0 0 1 1 / 2 ⎥⎦
Now, the above matrix is a unit matrix.

Therefore,
x = 5/4
146
Matrices
y = 7/4
z = 1/2
Now let us check whether these values satisfy the initial equations.
2y - 3z = 2(7/4) - 3(1/2) = (7/2) – (3/2) = 4/2 = 2
2x + z = 2(5/4) + (1/2) = (5/2) + (1/2) = 6/2 = 3
x - y + 3z = (5/4) - (7/4) + 3(1/2) = -(2/4) + (3/2) = 2/2 = 1
The values of x, y, z calculated using Gauss - Jordan elimination method satisfy the
equations in the example.
Thus, a manager can use Cramer’s rule, matrix inversion method or Gauss - Jordan
elimination method to solve a system of equations.
SUMMARY
Matrix is a set of ‘m × n’ numbers arranged in the form of rectangular array having

‘m’ rows and ‘n’ columns. It is called an m x n matrix. Matrices play an important
role in modern techniques of quantitative analysis of managerial decisions. Matrices
provide a compact way of representing a system of equations.
This chapter discussed about determinants, adjoint and inverse of a matrix. To the end
of the chapter we have discussed how managers can use matrix methods like Cramer’s
Rule, matrix inversion method and Gauss - Jordan elimination method to solve linear
equations related to various managerial problems. Thus, matrices play a vital role in a
manager’s decision making process.
147
Chapter 10
Interpolation and
Extrapolation
• Assumptions of Interpolation and Extrapolation
• Applications of Interpolation and Extrapolation
• Methods of Interpolation and Extrapolation
• Applications in Financial Analysis
Interpolation is a method of statistical estimation and the word literally means `making
insertions’. Let us consider a well-known situation which requires the use of
interpolation. Assume the census is conducted once in ten years. So, typically the exact
head count of the population of the country is available only for years 1921, 1931,
1941, 1951, 1961, 1971, 1981 and 1991. If we need to know population of the country
for any intermediary year, say, 1985, one logical approach would be to work forwards
from the population of 1981, by adding births and inflow of people into the country
and deducting deaths and outflow of people from the country during the period 1981 to
1985. Though, this approach would give us an accurate population figure for 1985, the
required data on births, deaths, inflow and outflow of people may be difficult to gather
and the cost of compiling such data may exceed the benefit that may accrue from the
exercise. The population figure for the year 1985, may be required to correlate the
demand of a product with the population and the inferences to be drawn may not be
highly sensitive to minor variations in the population figure. That is, while the data on
population in 1985 is required, a one hundred percent accurate figure is really not
required.
Interpolation is a statistical technique, which through a study of the time series of
known figures of population, say from 1921 to 1991, allows us to make an insertion of
the population figure for the year 1985.
This is usually accomplished with a few steps of calculation and the figure so obtained
is fairly correct as long as the assumptions underlying interpolation hold good.
Suppose, the same firm by studying the past correlation of demand for its products
with the population wants to predict the demand for a future year, say, 1997, through a
forecast of the population figure. This is possible through the use of the technique
called extrapolation.
Extrapolation allows us to forecast or anticipate a value for some future date.
What is the basic requirement for both interpolation and extrapolation to work? There
must exist a functional relationship between an independent variable and a dependent
variable and it must be required to estimate the value of the dependent variable
corresponding to a certain value of independent variable. This functional relationship
is usually developed on the basis of past data using a branch of statistics called
‘Regression Analysis.’
In the situations considered above, population is taken to be a function of time or years.
While time is the independent variable, population is the dependent variable. One more
point that requires emphasis is that both interpolation and extrapolation provides us
with only estimates of the dependent variable.
ASSUMPTIONS OF INTERPOLATION AND EXTRAPOLATION
The following are some of the assumptions of interpolation and extrapolation:

• The figures known are assumed to be a normal series, that is a series without any
violent, unexplained fluctuations in the values. The period intervening the points
of known data and the point for which an estimate is required is also considered to
be a normal period.
For instance, when we require the interpolation of the population figure for the
year 1985, we assume that between 1981 and 1985 there have been no cause for
sudden jumps or declines in population such as war, flood, famine, cyclone and
the like.
• It is also presumed that the increases and decreases take place only at a uniform
rate. For instance, if the population in 1981, was 700 million and in 1991, 850
million, it is assumed that the accretion to the population has taken place at a
steady growth rate.
149
APPLICATIONS OF INTERPOLATION AND EXTRAPOLATION

Interpolation and extrapolation are widely used by businessmen, administrators,
sociologists, economists and financial analysts. While interpolation helps in
completing the incomplete, lost or destroyed records, extrapolation is helpful in those
circumstances where forecasting and prediction are required. In financial analysis, the
technique of interpolation is widely used to find internal rate of return (IRR) of a
project, yield to maturity of a bond and in all investment decisions which require the
use of the present value and future value interest factor tables.
METHODS OF INTERPOLATION AND EXTRAPOLATION

Though, there are a variety of methods which can be used for interpolation, we will
discuss and illustrate only the graphical and linear approximation methods.
Graphical Method
As the name suggests, this method arrives at the solution after plotting a graph using
the given data. While drawing the graph on a natural scale, the independent variables
are marked along the horizontal line and corresponding dependent variables are
marked along vertical line. The points so plotted are joined by a smooth curve. For
interpolation, the value of the independent variable is marked at the appropriate place
on the X-axis and a line perpendicular to the X-axis is to intersect the curve. Again by
drawing a line perpendicular to the Y-axis from the point of intersection on the curve
to meet the Y-axis, the value of the dependent variable is ascertained by reading off the
value at the point of intersection with the Y-axis. Similarly, graphical method can be
used for extrapolation. The illustration given below demonstrates the use of this
method.
Example 10.1
The population of a town is as follows:
Year 1941 1951 1961 1971 1981 1991
Population (in 000’s) 24 30 40 54 78 104
Find the population for the years 1976 and 1992.
Figure 10.1: Graphical Method

120
100
Population in ‘000s
80
60
40
20
1941 51 61 71 81 91 Year
76 92
150
Solution:
Marking points on the X-axis, for the years 1976 and 1992, and dropping
perpendiculars, the population for the years can be interpolated as 65,000 and 125,000
respectively.
Linear Approximation Method

This is a rough and ready method of interpolation and is best used when the series
moves in predicted intervals. It can be applied to interpolate values in both ascending
and descending series. The method can be best described with the help of illustrations.
1. Use the value of the immediately preceding year as the base value or starting
value.
2. Ascertain whether the series is an ascending one or a descending one.
3. Find out the difference in values of the variable corresponding to the immediately
succeeding and preceding years or the year for which the value is to be
interpolated. We may connote this as upper limit minus lower limit.
4. Find out the time interval between the two known values. We may denote this as
ts – tp.
5. Find out the time interval between the immediately preceding year and the year for
which the value is to be interpolated. We may denote this as ti – tp.
6. Interpolate the value as follows for ascending series
Upper Limit − Lower Limit
Base value + × (ti − t p )
(t s − t p )
If the series is a descending series, the formula will be,

Lower limit − Upper Limit
Base value − × (ti − t p )
(t s − t p )
Example 10.2
The sales in units of a consumer durable are ascertained as follows:
Year 1986 1987 1988 1990
Sales Units (in ‘000s) 8 16 24 40
The records of sales for the year 1989 were accidentally lost in a fire. Find the sales for
the year 1989.
Solution:
The sales in this year could be interpolated by the following procedure:
1. The year immediately preceding 1989, is 1988. Hence 24,000 units sold in 1988 is
the base value.
2. In the illustration, since demand for the units is steadily increasing, we may
conclude that it is an ascending series.
3. Value corresponding to the immediately succeeding year (1990), 40,000.
Value corresponding to the immediately preceding year (1988), 24,000
Upper limit – Lower limit = 40,000 – 24,000= 16,000
4. In the above illustration, the time interval between 1988 and 1990 is 2 years.
5. Time interval between 1988 and 1989 is one year.
151
6. For the illustration, the sales for the year 1989 will be,
24000 + 40,000 - 24,000 ×1
2
= 24,000 + 8,000 = 32,000
Example 10.3
Year 1984 1985 1986 1987 1990
Advertising expenditure (Rs. in lakhs) 45 39 33 27 9
Solution:
The advertising expenditure incurred in 1989 can be interpolated as follows:

Base Value – Rs. 27 lakhs
Lower Limit – Rs. 27 lakhs
Upper Limit – Rs.9 lakhs
ts – tp = (1990 – 1987) = 3
ti – tp = (1989 – 1987) = 2
Advertising expenditure in 1989 = 27 − 27 − 9 × 2 = 27 − 12 = Rs. 15 lakh
3
APPLICATIONS IN FINANCIAL ANALYSIS

In financial analysis, interpolation is used widely in:
a. Determination of internal rate of return of a project.
b. Finding out the yield to maturity (YTM) of a bond or debenture and
c. Other situations where the time value of money is considered and interpolations
have to be made while using the present and future value tables.
It must be noted that in all the above, interpolation is used to insert figures in a series
other than a time series.
In financial analysis extrapolation is widely used for:
a. Forecasting future sales, cost and profit.
b. Long-term capital requirements.
c. Production of financial statements for financial institutions, banks, etc.
Example10. 4
The cash inflows of a project involving an initial outlay of Rs.22 lakhs is as follows.
Find internal rate of return.
Year Rs. in lakhs
1 10
2 10
3 6
4 3
Solution:
The internal rate of return is the rate at which the total value of discounted cash
outflows is exactly equal to the total of discounted cash inflows. The internal rate of
return of a project can be determined only through a process of trial and error.
152
To begin with, let us try the discount rate of 14%.

Using present value interest factor (PVIF) tables, the total of discounted cash inflows
will be,
(10 × 0.877) + (10 × 0.769) + (6 × 0.675) + (3 × 0.592)
= Rs.22.29 lacs.
Since this figure is higher than the initial outflow of Rs.22 lacs, we must discount at a
higher rate.
At r = 15%, the total of discounted cash inflows will be,
(10 × 0.870) + (10 × 0.756) + (6 × 0.658) + (3 × 0.572)
= Rs.21.92 lacs.
At the discount rate of 15%, the discounted cash inflows are slightly lower than Rs.22
lacs. It can be concluded that the internal rate of return must lie somewhere in between
14% and 15%. The technique of interpolation can be used to determine the exact rate
of return.
We now have a series of the following nature:
Rate % Discounted Cash Flows
(DCF)(Rs. in lakhs)
14 22.29
15 21.92
For an intermediary figure of Rs.22 lakhs of discounted cash flow we need to
interpolate the rate.
The linear approximation method may be used to interpolate. We know that when the
rate increases by 1%, the DCF falls from 22.29 to 21.92 or the descent in DCF for 1%
ascent in rate is (22.29 – 21.92). We also know that the interest rate must be higher
than 14%, but less than 15%.
At the exact rate, the descent must be (22.29 – 22.00). When descent is (22.29 –
21.92), the increase in rate is 1. For a descent of (22.29 – 22)
The increase in rate must be
22 . 29 − 22
×1
22 . 29 − 21 . 93
The internal rate of return =
22 . 29 − 22 = 14 + 0.805 = 14.805%
14 + ×1
22 . 29 − 21 . 93
Example 10.5
The market price of Rs.1,000 face value debenture is Rs.800. The coupon interest rate
is 9% and the debenture matures at the end of 8 years at which time it will be redeemed
at the face value. What is the yield to maturity (YTM)?
(The concept of yield to maturity is very similar to the concept of internal rate of
return. YTM is the rate at which the inflows from holding the debenture when
discounted will exactly equal the current market price of the debenture.)
Solution:
The inflows from holding the debenture are,
Years 1 to 8 Interest Rs.90 per year.
End of 8th year Principal Rs.1000
153
If `k’ is the YTM

8 90 1000
800 = ∑ +
i = 1 (1 + k )i (1 + k ) 8
At k = 14%, the right hand side of the above equation will be
= 90 × PVIFA (14, 8) + 1000 × PVIF (14, 8)
= (90 × 4.639) + (1000 × 0.351)
= Rs.768.51
At k = 13%, the right hand side of the equation will be
= 90 × PVIFA (13, 8) + 1000 × PVIF (13, 8)
= (90 × 4.799) + (1000 × 0.376)
= Rs.807.91
From the above, we may conclude that the YTM lies in between 13% and 14%. The
series we have is,
Rate DCF
% Rs.
13 807.91
14 768.51
When the descent is (807.91 – 768.51), the increase in rate is 1. For a descent of
(807.91 – 800), the increase in rate should be,
807 . 91 − 800
×1
807 . 91 − 768 . 51
The YTM of the bond = 13 + 807 . 91 − 800
× 1 = 13 + 0.20 = 13.2%
807 . 91 − 768 . 51
Example 10.6
A finance company offers to pay a lump sum of Rs.16, 000 at the end of 6 years to
investors who deposit Rs.2, 000 annually for 6 years. What is the implicit interest rate?
Solution:
We know that 16,000 = 2000 × FVIFA (k, 6)
Where FVIFA is the future value interest factor of an annuity, and k is the implicit rate.
FVIFA (k, 6) = 16,000/2000 = 8
If we refer to the FVIFA tables,
at k = 11%, FVIFA(11,6) = 7.913 and
at k = 12%, FVIFA(12,6) = 8.115
So, k must be greater than 11% but lower than 12%.
For an ascent of (8.115 – 7.913), the ascent in rate is 1. For a required ascent of (8.000
– 7.913), the ascent in rate is
8 . 000 − 7 . 913
×1
8 . 115 − 7 . 913
The implicit rate of interest = 11 + 8 . 000 − 7 . 913 × 1 = 11 + 0.43 = 11.43%
8 . 115 − 7 . 913
154
SUMMARY
Interpolation and extrapolation are statistical methods of estimation and forecasting.
Interpolation is a statistical technique, which through a study of the time series of
known figures of population allows us to make a data insertion between a given data
set. On the other hand, extrapolation allows us to forecast or anticipate a value for
some future date.
While interpolation helps in completing the incomplete, lost or destroyed records,
extrapolation is helpful in those circumstances where forecasting and prediction are
required. Businessmen, administrators, sociologists, economists and financial analysts
widely use interpolation and extrapolation. The two methods of interpolation are -
graphical method and the linear approximation method.
155
Appendix
Table 1: Areas under the Standard Normal Distribution between the Mean and
Positive Values of z
a 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0.0 0.0000 0.0040 0.0080 0.0120 0.0160 0.0199 0.0239 0.0279 0.0319 0.0359
0.1 0.0398 0.0438 0.0478 0.0517 0.0557 0.0596 0.0636 0.0675 0.0714 0.0753
0.2 0.0793 0.0832 0.0871 0.0910 0.0948 0.0987 0.1026 0.1064 0.1103 0.1141
0.3 0.1179 0.1217 0.1255 0.1293 0.1331 0.1368 0.1406 0.1443 0.1480 0.1517
0.4 0.1554 0.1591 0.1628 0.1664 0.1700 0.1736 0.1772 0.1808 0.1844 0.1879
0.5 0.1915 0.1950 0.1985 0.2019 0.2054 0.2088 0.2123 0.2157 0.2190 0.2224
0.6 0.2257 0.2291 0.2324 0.2357 0.2389 0.2422 0.2454 0.2486 0.2517 0.2549
0.7 0.2580 0.2611 0.2642 0.2673 0.2704 0.2734 0.2764 0.2794 0.2823 0.2852
0.8 0.2881 0.2910 0.2939 0.2967 0.2995 0.3023 0.3051 0.3078 0.3106 0.3133
0.9 0.3159 0.3186 0.3212 0.3238 0.3264 0.3289 0.3315 0.3340 0.3365 0.3389
1.0 0.3413 0.3438 0.3461 0.3485 0.3508 0.3531 0.3554 0.3577 0.3599 0.3621
1.1 0.3643 0.3665 0.3686 0.3708 0.3729 0.3749 0.3770 0.3790 0.3810 0.3830
1.2 0.3849 0.3869 0.3888 0.3907 0.3925 0.3944 0.3962 0.3980 0.3997 0.4015
1.3 0.4032 0.4049 0.4066 0.4082 0.4099 0.4115 0.4131 0.4147 0.4162 0.4177
1.4 0.4192 0.4207 0.4222 0.4236 0.4251 0.4265 0.4279 0.4292 0.4306 0.4319
1.5 0.4332 0.4345 0.4357 0.4370 0.4382 0.4394 0.4406 0.4418 0.4429 0.4441
1.6 0.4452 0.4463 0.4474 0.4484 0.4495 0.4505 0.4515 0.4525 0.4535 0.4545
1.7 0.4554 0.4564 0.4573 0.4582 0.4591 0.4599 0.4608 0.4616 0.4625 0.4633
1.8 0.4641 0.4649 0.4656 0.4664 0.4671 0.4678 0.4686 0.4693 0.4699 0.4706
1.9 0.4713 0.4719 0.4726 0.4732 0.4738 0.4744 0.4750 0.4756 0.4761 0.4767
2.0 0.4772 0.4778 0.4783 0.4788 0.4793 0.4798 0.4803 0.4808 0.4812 0.4817
2.1 0.4821 0.4826 0.4830 0.4834 0.4838 0.4842 0.4846 0.4850 0.4854 0.4857
2.2 0.4861 0.4864 0.4868 0.4871 0.4875 0.4878 0.4881 0.4884 0.4887 0.4890
2.3 0.4893 0.4896 0.4898 0.4901 0.4904 0.4906 0.4909 0.4911 0.4913 0.4916
2.4 0.4918 0.4920 0.4922 0.4925 0.4927 0.4929 0.4931 0.4932 0.4934 0.4936
2.5 0.4938 0.4940 0.4941 0.4943 0.4945 0.4946 0.4948 0.4949 0.4951 0.4952
2.6 0.4953 0.4955 0.4956 0.4957 0.4959 0.4960 0.4961 0.4962 0.4963 0.4964
2.7 0.4965 0.4966 0.4967 0.4968 0.4969 0.4970 0.4971 0.4972 0.4973 0.4974
2.8 0.4974 0.4975 0.4976 0.4977 0.4977 0.4978 0.4979 0.4979 0.4980 0.4981
2.9 0.4981 0.4982 0.4982 0.4983 0.4984 0.4984 0.4985 0.4985 0.4986 0.4986
3.0 0.4987 0.4987 0.4987 0.4988 0.4988 0.4989 0.4989 0.4989 0.4990 0.4990

Foundations of Quantitative Methods

Uploaded by

Copyright:

Available Formats

Foundations of Quantitative Methods

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

Foundations of Quantitative Methods

Uploaded by

Copyright:

Available Formats

Foundations of

Ref. No. FQM/A – 03 2K5 45

Chapter 2 Basics of Mathematics 5

Chapter 4 Arranging Data 47

Chapter 5 Measures of Central Tendency 61

Chapter 6 Measures of Dispersion 79

Chapter 7 Index Numbers 93

Chapter 8 Introduction to Probability and Probability Distributions 106

Chapter 9 Matrices 126

Chapter 10 Interpolation and Extrapolation 148

Managers need to analyze environmental factors on a continuous basis, determine the

EVOLUTION OF QUANTITATIVE METHODS

CLASSIFICATION OF QUANTITATIVE METHODS

standard deviation, coefficient of correlation etc., the methodologies of analyzing and

ADVANTAGES OF QUANTITATIVE METHODS

LIMITATIONS OF QUANTITATIVE METHODS

Though quantitative methods are applied to an increasingly large number of business

As the process of decision-making is becoming increasingly difficult, managers can no

086 = 0 x 102 + 8 x 101 + 6 x 100

The Binary Number System

Example for carryover in Decimal system

(b) (111000)2 = 1×25 + 1×24 +1×23 + 0×22 + 0×21 + 0×20

(c) (1100101)2 = 1×26 + 1×25 + 0×24 + 0×23 + 1×22 + 0×21 + 1×20

Conversion of fractional decimal to binary number

Octal Number System

Conversion of fractional decimal to octal

= 0.875 + 0.078125 + 0.00390625

PERMUTATIONS AND COMBINATIONS

n and r are positive integers and n ≥ r.

Find the logarithm of 64 5 4 to base 4 2

∴ (32. 31/2)x = 34. 31/4

To Find the Logarithm of a Product

To Find the Logarithm of a Fraction

To Find the Logarithm of a Number Raised to Any Power, Integer or Fraction

Sign of the Logarithm

The following are inequalities in logarithms:

= log 3 − 2 log 2 − 2 log3 + 2 log2 + 4 log 3 – 3 log 2 + 3 log 2 − 3 log 3

Therefore, the nth term in an A.P. is given by

To find the sum of number of terms in arithmetical progression

The sum to n terms of an A.P. is given by

Substituting the value of d in equation (A), we get

Also given that

Therefore the nth term in a G.P. is given by

If b/a = c/b i.e if 1 - b/a = 1 - c/b

For a layman, “function” indicates a relationship among objects. A function provides a

Functions of Many Variables

Figure 2.1: Linear Functions

Equation of a straight line

By using the above formula we get,

System of Linear Equations

We must make Cell (2, 1) = 0. This is done by adding (2*Row 1) to Row 2.

This chapter intends to provide the student a fundamental knowledge of mathematics.

The value of f(x) becomes larger and larger, as x approaches 1.

( x 2 + 1)(6 x 2 + 1) − (2x 3 + x + 3)(2x )

Since, y depends on z, and z in turn depends on x, y is also a function of x.

Total Revenue = 100 MR = 100 – 20q

i. U (R) = aR – bR2 where a and b are constants

Data is a collection of related observations, facts or figures. Managers collect facts

Examples of Raw Data

Methods of Collecting Data