Algebra and Calculus PDF
Algebra and Calculus PDF
Algebra and Calculus PDF
Study Book
Written by
Chris Harman and Patricia Cretchley
Faculty of Sciences
The University of Southern Queensland
ii
c
The
University of Southern Queensland, 2007.
Published by
The University of Southern Queensland
Toowoomba Qld 4350
Australia
http://www.usq.edu.au
Copyrighted materials reproduced herein are used under the provisions of the
Copyright Act 1968 as amended, or as a result of application to the copyright
owner.
No part of this publication may be reproduced, stored in a retrieval system or
transmitted in any form or by any means electronic, mechanical, photocopying,
recording or otherwise without prior permission.
Produced using LATEX in the USQ style.
Preface
Welcome to Algebra & Calculus I! We look forward to meeting you, introducing you to a wonderful wide range of mathematical techniques and
applications, and assisting you to build academic and professional skills and
understanding via these studies!
You will study two modules, Algebra and Calculus, in parallel, and this
Study Book is intended to direct your path through the prescribed texts,
readings and activities for each module. Use it in conjunction with your
Introductory Book which has the weekly Study Schedule.
Reading your Study Book and texts is just the start! You will be required to
develop and demonstrate a broad range of skills. Your goal will be to apply
the course concepts and processes to both new and familiar situations, and
to communicate your methods and results using the appropriate terminology
and notation.
The assignment tasks alone will offer hopelessly inadequate practice. You
will need to do a wide selection of exercises to build your experience, gain
confidence, and learn to work quickly.
Besides this Study Book, the prescribed texts are
ELEMENTARY LINEAR ALGEBRA, 5th Edition (2004), by Larson, Edwards & Falvo, published by Houghton Mifflin; and
CALCULUS: Concepts & Contexts, 3rd Edition (2004 or 2005), by James
Stewart, published by Brooks/Cole.
iii
iv
Table of Contents
0.1
0.2
11
0.3
Vectors in Space . . . . . . . . . . . . . . . . . . . . . . . . .
17
0.4
18
0.5
20
0.6
Summary: Vectors . . . . . . . . . . . . . . . . . . . . . . . .
28
29
1.1
. . . . . . . . .
30
1.2
34
1.3
36
1.4
37
2 Matrices
2.1
39
40
vi
Table of Contents
2.2
41
2.3
42
2.4
Elementary Matrices . . . . . . . . . . . . . . . . . . . . . . .
43
2.5
43
2.6
Summary: Matrices . . . . . . . . . . . . . . . . . . . . . . .
44
3 Determinants
45
3.1
46
3.2
48
3.3
Properties of Determinants . . . . . . . . . . . . . . . . . . .
49
3.4
50
4 Complex Numbers
II
51
4.1
52
4.2
. . . . . . . .
53
4.3
54
4.4
56
4.5
Solving Equations . . . . . . . . . . . . . . . . . . . . . . . .
62
4.6
68
4.7
73
CLEVER CALCULUS
75
79
1.1
80
1.2
81
1.3
83
1.4
84
vii
Table of Contents
1.5
Exponential Functions . . . . . . . . . . . . . . . . . . . . . .
84
1.6
85
. . . . . . . . . . . . . . .
87
2.1
88
2.2
88
2.3
89
2.4
Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
90
2.5
91
2.6
91
2.7
Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . .
92
2.8
93
2.9
95
3 Differentiation Rules
97
3.1
98
3.2
99
3.3
3.4
3.5
3.6
3.7
4 Applications of Differentiation
105
4.1
4.2
4.3
4.4
. . . . . . . . . . . . . . . . 106
viii
Table of Contents
4.5
4.6
4.7
4.8
4.9
Antiderivatives . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5 Integrals
111
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
6 Applications of Integration
121
6.1
6.2
Volumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
6.3
6.4
6.5
125
Module
POWERFUL
AND
FASCINATING
ALGEBRA
Overview
This module has powerful, wide-reaching and important applications in most
fields of endeavour! Browse overleaf, to see applications of your Linear
Algebra to computer graphics, cryptography, computed tomography to gain
images of the human body, CAT scans, electrical networks, traffic networks,
forest management, population growth, genetics, temperature distributions,
chaos theory, and games of strategy, to list but a few! Some elementary
applications will be developed in this course; many more will arise in your
further studies.
Two and three-dimensional geometric Vectors were developed to represent
and study physical phenomena and motion in real-world applications. We
look at their geometric and algebraic properties, see how useful these are for
describing motion, and how readily and easily they can be used to describe
the properties and equations of lines and planes in space. The properties we
develop here extend very easily to vectors of more than 3 dimensions, and so
in Algebra & Calculus II we will apply them again to vectors of n-dimensions
in a host of further applications!
Large data sets and big systems of linear equations arise with astonishing
frequency in important scientific, industrial and economic applications. We
will see how to use matrices to store and manipulate data, and to manage
and solve systems of linear equations. We will characterise some important
types of matrices, and gain expertise with the different types of solutions
that emerge from linear systems.
Complex numbers are a brief and fascinating finale. While they arose historically out of attempts to create square roots of negative numbers, their
geometric and algebraic properties have yielded remarkably practical applications. We look briefly at how complex numbers and their functions are
commonly used to analyse real-world phenomena and solve equations.
The title Algebra really is a misnomer for this module, because in parallel we
develop the vitally important Geometry of vectors, matrices and complex
4
numbers. You will see how fruitful all are for applications, learn how to
exploit them, and become aware of strong links between them.
Module Objectives
By the end of this Module, you should recognise a host of areas of application
of vectors, matrices and complex numbers, be aware of their power and
beauty, and be able to exploit them in other areas and disciplines, as well
as in Algebra & Calculus II.
It is vitally important that you understand the links between these concepts.
Always read with pencil in hand, and confirm the steps as you go. Visualising
is a powerful aid: draw diagrams wherever you can!
Use technology where encouraged to do so! Its fun and helpful, and the
weekly tutorial/laboratory tasks and the assignments will help you develop
the necessary skills with our careful full support!
At the end of each Chapter you should
have mastered the concepts covered
be able to link their algebra and geometry
be competent with the processes, terminology, and notation
have mastered a wide range of exercises
be able to communicate your reasoning well, using correct language
and symbols
be able to apply the concepts in a range of contexts, new and familiar
At the end of this Module, you should
see links between vectors, matrices and complex numbers
recognise contexts in which they are valuable
be able to manipulate and apply these concepts competently
be able to demonstrate and communicate those skills appropriately
5
Chapter
Preliminary Work on
Vectors in the Plane
and in Space
Now what are vectors? And how do they differ from scalars?
Many quantities are fully characterised once their magnitude is known: eg
mass, length, volume, speed. They are known as scalar quantities.
Others are only fully specified when we give their direction as well: eg
the motion of a boat (speed and direction), wind velocity (strength and
direction), forces in a force-field, fluid flow. Vectors are quantities that
require more than one number for their specification.
There are many kinds of vectors. In this course we study geometric vectors
in 2 and 3 dimensions. We use their magnitude and direction to represent
movement or motion in the plane or in space, and we use their algebraic and
geometric properties to describe lines and planes in 3-dimensional space. In
Algebra & Calculus II, we generalise these definitions to vectors of greater
dimension.
This introduction to geometric vectors includes:
geometric representation as an arrow
coordinate representation as (x, y) or (x, y, z)
addition and subtraction, to find resultants
use of vectors to describe motion
dot product to evaluate lengths and angles
dot product as a test for perpendicularity
cross product to create a perpendicular vector
cross product to describe some areas and volumes
use of vectors to find equations of lines and planes in space
0.1
Reading
Appendix A: study section 3.1, from Elementary Linear Algebra, 4th Edition, by Grossman. Be sure Table 3.1 makes
sense.
Also read Stewart: sections 9.1 and 9.2. This will be particularly helpful when you reach 3-dimensional vectors in Section
0.3 of this Module.
You may use any of the following notations for vectors: each has
its strengths: (a, b) = ha, bi = a~i + b~j.
Stewart uses a different kind of bracket - possibly better to not use that
notation at this stage.
Be sure you have mastered the following:
representing a vector geometrically as an arrow or directed line segment, and algebraically as an ordered pair in round or pointed brackets;
finding its magnitude (length) from components, using Pythagoras;
finding its direction angle: it helps to sketch the vector so as to choose
the correct quadrant angle;
finding scalar multiples geometrically and algebraically;
adding or subtracting algebraically, and geometrically using a triangle
or parallelogram;
defining ~i and ~j and expressing vector (a, b) as a~i + b~j.
Note that if ~u and ~v form adjacent sides of a parallelogram, then their sum
and difference form the two diagonals. Show this on a sketch.
Be sure you can distinguish between the directions of ~u ~v and ~v ~u.
Note: Grossman and Stewart use single bars to denote the magnitude of
~v . Some texts use double bars ||~v || to emphasise it is not the absolute value
of a scalar quantity. Use the notation you favour, but it is important to
recognise and be familiar with both.
10
Activity
Note also the following rules for vector algebra. Check geometrically that each rule seems reasonable. Try some examples to get a feeling for why the rule works, but you must
use general vectors to prove the rules algebraically: ie of the
form ~a = (a1 , a2 ), ~b = (b1 , b2 ), in R2 , say.
For any vectors ~a, ~b and ~c, and scalars (numbers) and :
~ = ~a
~a + O
~ (note 0 is a scalar and O
~ a vector, so O
~ = (0, 0) in R2 );
0~a = O
~a + ~b = ~b + ~a
(~a + ~b) + ~c = ~a + (~b + ~c)
(~a + ~b) = ~a + ~b
( + )~a = ~a + ~a
()~a = (~a)
To make any vector ~v into a unit vector, note that we can simply
~v
~v
or ||v||
.
divide it (ie each of its components) by its own length: |v|
Activity
Do Grossman, Problems 3.1 of Appendix A, odd numbers
1-35 (31 optional), and as many others as you can manage.
Note that Grossman gives quite full worked solutions for the
odd numbers. These are a valuable aid.
Use examples and exercises from Stewarts Section 9.2 to
support your study, if you find that helps.
0.2
11
Addition and subtraction of vectors are done componentwise, and so is multiplication by a scalar. But we will not define the product of vectors (u1 , u2 )
and (v1 , v2 ) componentwise as (u1 v1 , u2 v2 ) here, because this definition turns
out to be less useful in the applications we need now, than the so-called dot
and cross products.
We study cross or vector product in Section 0.4.
The dot or scalar product of two vectors,
~u ~v = u1 v1 + u2 v2
is so-called because it gives a scalar from two vectors.
Note the most important property:
~u ~v = |~u| |~v | cos .
It is used
for finding the angle between ~u and ~v , and
to give a test for perpendicularity: two vectors are perpendicular if
and only if their dot product is 0.
Reading
Study Appendix A, Grossman, Section 3.2 in two parts: first
scalar (or dot) products, pp 164mid 168. Continue with
projections when you have mastered dot products.
12
Activity
Do Appendix A, Problems 3.2, odd numbers,1-19, to ensure
mastery.
Note Q 35 which gives a simple test for parallelness:
two non-zero vectors are parallel iff (if and only if ) one is a
scalar multiple of the other:
for example (2, 3) and (6, 9) are parallel
because (6, 9) = 3(2, 3).
Sketch to check. Note that the negative multiple tells us that
they point in opposite directions.
13
14
Activity
Stewart: study section 9.2 and do Exercises 1, 3, 5, 23, 24,
25.
Note on Exercise 24 above:
Draw the two vectors for the wind and the plane. To add them, attach the
tail of the planes vector to the head of the winds vector. The resultant
course is the third side of the triangle, where the base angle is 45+60=105
degrees. You can then use the cosine rule to find this third side. Then you
can use the sine rule to find the angles of the triangle.
15
We say ~u can be decomposed (or resolved) into the sum of those two vector
components, one parallel to ~v and one perpendicular to ~v .
Scalar component:
If is the angle between ~u and ~v , then the definition of cosine tells us that
the vector projv ~u has length |~u| cos .
But cos may be positive or negative, so |~u| cos is a signed length expressing
how far ~u extends in direction ~v .
We call |~u| cos the scalar component of ~u in the direction of ~v .
Note that dividing by |~v | in ~u ~v = |~u||~v | cos , gives
Scalar component = |~u| cos =
~u ~v
|~v |
~
u~v
|~v |
or |~u| cos ,
~v
|~v | .
~
u~v
|~v |
We conclude that
~v
projv ~u = |~u| cos
=
|~v |
~u ~v
|~v |
~v
=
|~v |
~u ~v
~v
|~v |2
16
Activity
You may find it helpful now to study Projections in Stewart:
section 9.3.
Then do Appendix A, Grossman: problems 3.2, odd numbers
1-19, 21-27, 33, 35 and 39. Master more, if you have time.
0.3
17
Vectors in Space
In this section, 2-dimensional geometric vectors in the plane, R2 , are extended to 3-dimensional space, R3 , by the inclusion of a third component in
the z-direction. The rules and definitions already established for R2 extend
easily, with a little checking.
Reading
Read Appendix A, Grossman, Section 3.3.
Take the emphasis off p 177, 178, direction cosines. Example
5 is optional.
Also read Stewart Sec 9.1.
Activity
Do Appendix A, Grossman problems 3.3, odd numbers 1-9,
21, 23, odds 27-39. Do as many others as you can manage.
18
Reading
Study Appendix A Section 3.4. Omit mid-187 to mid 188.
Also omit Theorem 1 and Example 2 if you are not familiar
with determinants: we study them in Chapter 3.
Use Sec 9.4 from Stewart as back-up.
Finding cross products:
Be careful: note that the middle component in the definition on p 183 has
the opposite order of subtraction to the other two components, causing a
change of sign.
If you know how to expand a determinant Theorem 1 offers a method of
calculation. But even if you have never seen a determinant, it is easy to
proceed as follows. Simply write the components of the second vector directly beneath the first: eg for the vectors in Example 1, p 184, write
1 1
2
2
3 4
use a cross
1
2
19
Activity
Do Problems 3.4, Q 3, 5, 9, 17, 19, 21, 23, 35, 29, 31, 33 37.
Do as many others as you can manage.
Note that you can also use cross product to find the area of
a triangle with adjacent sides ~u and ~v : the area is half that
of the parallelogram they define.
Activity
Browse through the approach to properties of vectors and
their uses in the Vectors Chapter in Stewart: sections 9.1 to
9.4.
Note the sections on dot and cross product, in particular.
20
Reading
The notes below replace those on the bottom of Grossman,
p 193194 of Appendix A, Section 3.5.
Stewart Section 9.5 provides further back-up.
The Equation of a Line:
To specify a particular line, in the plane or in space, it is sufficient to know
just one point on it, as long as we know its direction.
Suppose we have a line, firstly in the plane for simplicity, with y-intercept
3, and slope 4.
Sketch this line here, and convince yourself that because the slope is 4/1, the
vector (1,4) is parallel to it.
Notice how we can travel from the origin to any point (x, y) on the line,
using the known point (0, 3) (the y-intercept) and the slope or direction
vector, (1, 4):
Firstly, to reach the line, we can travel along the position vector (0, 3)
to get to the only known point on the line.
21
Then to move from this point to any other point on the line, say
higher, we need to add further travel in the direction of the line, ie in
the direction of the slope or direction vector (1, 4).
How far should we travel in the direction of (1, 4)? We want to get to any
point on the line.
If we travel exactly 1 such direction vector, we get to the point (0, 3) +
(1, 4), ie (1, 7). Confirm this graphically: travel to (0, 3) first, then
travel the vector (1, 4) which is parallel to the line.
If we travel 2 slope vectors (1, 4) instead, after going to (0, 3), we reach
the point (0, 3) + 2(1, 4) ie (2, 11).
Using big positive multiples of the direction vector will take us to
points much further up the line.
To reach points further down the line from (0, 3), we need to travel in
the opposite direction, ie in the direction of (1, 4) = (1, 4). Using
ever bigger negative multiples of the direction vector will take us to
points further on the other side of (0, 3).
To conclude: as long as we allow t to have all possible positive and negative
values, the sum (0, 3) + t(1, 4) will take us to all possible points (x, y) on the
line. This means that the line comprises all possible points of the form
(x, y) = (0, 3) + t(1, 4), where t is any real number.
There is nothing special about the particular point and direction vector we
used: we can travel to any fixed point (x1 , y1 ) on the line, then simply add
multiples t of any direction vector (a, b).
For example, we could have used any multiple of (1, 4) as the direction vector
in the example above: that would still give a vector parallel to the line. If we
used (2, 8) instead, then to reach a particular point, we would need exactly
half the multiple required if we were using direction vector (1, 4). So the
line is also defined by the vector equation
(x, y) = (0, 3) + s(2, 8), s any real number.
In general, the vector equation of a line in R2 through given point
(x1 , y1 ), and parallel to the vector (a, b), has form
(x, y) = (x1 , y1 ) + t(a, b), t any real number.
22
23
Activity
Study Examples 1, 2 and 3, on p 195 of Grossman in
Appendix A. Take the emphasis off Examples 4, 5 and 6.
Also study Stewart: section 9.5 Examples 1, 2.
Then do Problems 3.5, odd numbers 1-3, 9-19, and 23-25.
We need vector or parametric forms later, rather than
symmetric form, so aim at finding vector form first, and be
able to change to parametric form.
Appendix A uses ~i, ~j and ~k notation, but component notation is often quicker and clearer for lines. For example,
the equation on line 6 of p 195 can be written simply as
~ = (2, 1, 6) + t(1, 2, 8).
OR
24
Reading
The notes below are a replacement for most of p 197 of App
A, Section 3.5. They are very important.
Study them carefully before reading pp 198 - mid 201. Pp
201 and 202 may be omitted, but you may want to come
back to Example 10 when you have learned to row-reduce a
matrix.
Stewart Section 9.5 provides further discussion, but you need
not study Examples 7, 9 and 10.
We seek a property that will define that plane by distinguishing all the points
on it from other points in space.
In R3 , a lines direction is characterised by a direction vector - ie any vector
parallel to it.
Consider what we regard to be a plane in space: a flat sheet extending
without bound in any direction. Many vectors with different directions are
parallel to it, and hence its direction cannot be characterised by a parallel
vector.
But note that there is only one direction perpendicular to a plane.
Imagine all vectors perpendicular to a given plane: note that they are parallel, though some have opposite directions.
We say all such vectors are normal to the plane (normal meaning perpendicular) and we call any one of them a normal for that plane.
A plane has many normals: if you have one, any positive or negative
multiple will also be normal to that plane.
Parallel planes have parallel normals. Draw some to see this.
Planes that are not parallel will have normals that are not parallel.
Convince yourself: draw some.
25
26
This are all different forms of the equation for the plane, and give us different
ways of finding it. Notice in particular the easy method:
(variable point) (normal) = (fixed point) (normal).
Vector ~n = (a, b, c), so this gives
(x, y, z) (a, b, c) = (x0 , y0 , z0 ) (a, b, c),
and taking dot products gives
ax + by + cz = ax0 + by0 + cz0 .
The LHS cannot be simplified because x, y and z are variables. But the
RHS will reduce to an answer, d, say.
So in general the equation of a plane has form ax + by + cz = d.
Note also that in this form, the coefficients (a, b, c) give a normal to the
plane.
Example:
1. The plane that passes through the point (1, 2, 3) in such a way that
it is normal to the vector (2, 1, 4), has equation
(x, y, z) (2, 1, 4) = (1, 2, 3) (2, 1, 4.)
Evaluating the dot products gives
2x y + 4z = 2 2 12, or simply 2x y + 4z = 12.
To find points (x, y, z) on this plane, we can substitute values for any
two of the variables, and work out the third. For example, substituting
easy values x = 0 and y = 0, we find 4z = 12, so that z = 3. The
point (0, 0, 3) therefore lies on that plane.
Verify by substitution that the point (1, 1, 13
4 ) also lies on the plane,
and find another point.
Also see Example 7, Appendix A, p 198, and the examples labelled i,
ii and iii at the bottom of p 198 and the top of p 199.
2. Sometimes the normal is not given, but can be gleaned from other
information: eg, if the plane is known to be parallel to another plane
with given equation, then one can use the known ones normal (ie its
coefficients of x, y and z) as a normal for the required plane;
27
Reading
Study any material you have not yet covered in Grossman,
pp 198-200 in Appendix A.
The material on pp 201-202 is optional and may be omitted,
but you may want to return to Example 10 after Sections 1.1
and 1.2.
Be sure that you have mastered the following:
the general form of the equation of a plane is ax + by + cz = d;
to find points on a plane with a given equation, we can substitute any
values we like for two of the variables, and w ork out the corresponding
value for the third;
given an equation of a plane of that form, the coefficients (a, b, c) give
a vector normal to that plane;
parallel planes have parallel normals (ie scalar multiples of each other);
perpendicular planes have normals that are perpendicular (ie have dot
product 0).
Activity
Do Grossman, Problems 3.5 in Appendix A, odd numbers
29-49, 50, 51, 53 and 55. Do any more you can manage.
28
Activity
Note that lines in R2 have many similarities to planes in
R3 . Plot roughly to scale the lines in R2 with equations
2x + y = 5 and 3x y = 2. See that vector (2, 1) is normal
(perpendicular) to the first line, and (3, 1) is normal to the
second. Show that the vector (a, b) is normal to the line with
equation ax + by = c, by showing that this line has slope
a/b, and using the fact that perpendicular lines have slopes
that multiply to give -1.
Chapter
Systems of Linear
Equations
29
30
31
Draw some of these possibilities, and see if you can think of others.
Reading
Study Section 1.1 of Larson et al. All of the material is
important.
Be sure that you have mastered the following:
be able to tell whether any given equation is linear or not
know the general form of a linear equation
know what is meant by a system of linear equations and its solutions,
if any
know that consistent means having at least 1 solution, inconsistent
means having no solutions at all
be able to express one variable in terms of all the others in a linear
equation
know what is meant by the terms free variable and parameter
be able to find particular solutions for the other variables by substituting arbitrary values for the parameters
Note that to solve a system of linear equations, we try change the system
into a simpler one that retains the same solutions. Note that we use the
elementary row operations to do that because they do not spoil the solutions.
Know exactly what the three elementary row operations are.
Aim at obtaining a triangular structure, called row-echelon form, or
else a diagonal form
If necessary, use back-substitution to find values for the variables
Note also that solving a system of linear equations of any size and in any
number of variables, leads to only three possibilities:
exactly 1 solution,
infinitely many solutions, or
32
Example 4 makes the last point clear for 2 equations in 2 variables. Solutions
are the points of intersection, if any, of 2 lines. It is impossible to find a
solution which comprises 2 points, or 10 points, say.
We find
exactly 1 point of intersection (when they cross), or
infinitely many (when the lines coincide), or
none at all (when the lines are parallel).
Activity
Confirm that for the intersection of planes in space, we still
only get three possibilities: one point, or none, or infinitely
many. Draw rough sketches of all the possible ways they can
intersect. Start with the intersection of just 2 planes. Then
try 3 planes. Examples 7 and 8 illustrate some possibilities.
Confirm that there will be
exactly 1 point of intersection, if the third plane cuts
through the line of intersection of the first two
none at all if the third plane misses the line of intersection of the other two: ie the planes intersect only
two at a time as though they frame an A
infinitely many points of intersection along a line if the
planes meet like the pages of a book
infinitely many along a plane if the 3 planes all coincide
none at all if all of the planes are parallel
none at all even if only two are parallel, because those
two have no points of intersection
Think about the effect of introducing a fourth plane too.
Activity
Section 1.1 Exercises, a selection of odd numbers, 1-25,
35-37, 43.
Also do Ed 4: 51-59, and 73;
or Ed 5: 51-64, 77.
33
34
35
that a homogeneous system (ie one with RHSs all 0s) always has
at least one solution, when all the variables are 0s (called the trivial solution) but that it may also have others, infinitely many of the
parametric type. So homogeneous systems are always consistent.
Activity
Look again at Example 8. Imagine that the variables were
x, y and z. Then the two equations represent planes, and
solving them simultaneously will find their point/s of intersection,i f any. Because they are not parallel, we expect them
to meet along a line. Show that the solution found there can
be written in the form (x, y, z) = (2, 1, 0) + t(5, 3, 1) and
hence deduce that this is the equation of the line of intersection of the planes.
Then do Section 1.2 Exercises, odd numbers 1, 5, 9, 15-21.
Also do Ed 4: 29-43, 59;
or Ed 5: 29-45, 61.
36
Reading
Study Larson Examples 1 to 4, on fitting polynomials to
points. Then study at least one type of network: either
Example 5, or else electrical circuits as in Examples 6 and 7.
Also peruse, for interest, the reading on CAT scans and Computed Tomography in Section 11.13 of Appendix C, to see
how systems of linear equations arise when we use these powerful techniques gain images of the human body. Iterative
techniques are useful to treat the over-determined systems of
equations that arise in such applications.
Activity
Do Section 1.3 Exercises, 1, 5, 7, 11 (If you have Ed 5, also
do Q15).
Also do two network problems of the type you have chosen:
from Ed 4, 19-26 (or Ed 5, 21-28).
1.4
37
Activity
Check through the summary at the end of Section 1. You
can find another on Systems of Linear Equations in the
coloured pages of the Study Guide Appendix A.
Attempt some of the Review Exercises in Larson et al, and
look at the Projects.
38
Chapter
Matrices
Matrices are two-dimensional arrays that offer a concise way to store essential information for any type of application. In this chapter, we develop
ways to manipulate matrices and do matrix algebra that prove useful for
managing and analysing many different types of data-sets. We will also find
links between matrices and vectors and their algebra: the rows or columns
of a matrix can be viewed as vectors, and a vector can be viewed as a row
or column matrix.
39
40
Chapter 2. Matrices
a b c
1 2
(1a + 3b + 5c) (2a + 4b + 6c)
d e f 3 4 = (1d + 3e + 5f ) (2d + 4e + 6f ) .
g h i
5 6
(1g + 3h + 5i) (2g + 4h + 6i)
Note that the inner dimensions of the matrices must agree, for the dot
products to work, and the outer dimensions give the size of the product
matrix: (n m)(m k) = (n k).
Note that AB need not equal BA. As a result, some of the algebraic rules
we are accustomed to do not hold: for example, we will find that when we
expand (A+B)(AB) according to the distributive rule in the next section,
the terms BA AB that arise may not give O, so these factors may not
yield the usual difference of squares result, A2 B 2 .
Note also that matrix multiplication is defined so as to be useful for applications, especially for systems of linear equations. Be sure you have mastered
being able to transform a linear system into a single matrix equation: see p
49/50 of Ed 4 (or p 52/3 of Ed 5).
Activity
Section 2.1 Exercises:
do Ed 4: odd numbers, 3, 7-11, 15, 19-31, 37, 50, 51, 52;
or Ed 5: odd numbers 3, 7-11, 15, 21-33, 39, 54, 57.
Attempt more if you have time.
2.2
41
Reading
Study Section 2.2 in Larson et al. An easy section, but very
important: be careful of the algebraic differences referred to
above, for example those illustrated by Examples 4 and 5,
and discussed in the notes between.
Note also that we can write the dot product of two vectors as a matrix
multiplication:
if ~u and ~v are both row vectors of equal length, then ~u ~v can be written as
the matrix product ~u~v T . Explain why it is necessary to transpose ~v in order
to write this matrix mltiplication.
Activity
Section 2.2 Exercises 3, 5, 9, 13, 17, 18, 19, 22, 23, 26, 27.
Also do Ed 4: 29, 36, 39, 53;
or else Ed 5: 29 (use technology), 31, 38, 41, 55.
42
Chapter 2. Matrices
Reading
Study Section 2.3 of Larson et al.
Note
the definition of the inverse of a matrix;
that when the inverse exists it is unique;
a method of finding inverses: we can row-reduce [A : I];
the form of the inverse of a 2 2 matrix at the bottom of p 69 of Ed
4 ( or p 73 of Ed 5). You may want it often enough to make it worth
remembering. Verify it by multiplying it by the original to get I.
"
a b
Notice that the inverse of
cannot exist if ad bc = 0. This number
c d
ad bc is called the determinant of the 2 2 matrix. We will define determinants fully later, so that they do indeed determine whether or not A has
an inverse.
Activity
Master Section 2.3 Exercises, odds 1-19, 21, 23 with technology, and 27.
Also do Ed 4: 29, 35, 37, 42, 43, 45, 46, 49, 50;
or else Ed 5: 29 (with tech) 31, 37, 39, 44, 45, 47, 48, 51, 52.
2.4
43
Elementary Matrices
2.5
There are many ways in which matrices can be used to facilitate the storing,
processing, transmission, and analysis of data. This section introduces just
four of these:
stochastic matrices of probabilities, used for prediction;
coding matrices, used for cryptography;
input-output matrices, used in economics, for industrial output;
least squares regression, which is covered in Algebra & Calculus II.
Once more you have a choice: study one of the first three applications listed
above.
Reading
Look through Section 2.5 of Larson et al: study either stochastic matrices or cryptography or Leontief input- output
models.
Browse through Sections 11.7 to 11.12 in Appendix C, to see
matrix operations and methods in use for some other applications: computer graphics, forest management, and equilibrium temperature distributions.
Activity
Do at least two exercises from Section 2.5, on the type of
application you selected. Write a summary paragraph of at
least 150 word on how matrices are used in another application that is of particular interest to you.
44
Chapter 2. Matrices
Activity
Check the Chapter Summary in Larson et al. Grossmans
summary on Matrices is also included in the coloured pages
of Appendix A in the Study Guide.
Activity
You may want to attempt some of the Review Exercises. Also
look over the types of questions asked in the Project 1.
Chapter
Determinants
We know how to clear the factor 3 from the LHS of the equation 3x = 5. We
simply multiply both sides by its multiplicative inverse or reciprocal, 1/3.
We also say we divide by 3.
We can clear any factor except 0 in that way. 0 is the only real number that
does not have a multiplicative inverse.
Let us apply a similar argument to matrix algebra: suppose we have matrix
equation AX = B. Some square matrices A have multiplicative inverses,
A1 , but many do not. This means that we cannot always get rid of a
matrix factor A in a matrix equation.
That is why we do not define matrix division: it simply cant always be done!
To be able to distinguish between those matrices that have inverses and
those that do not, we now define a number called the determinant of a
matrix, so that
when the determinant of the matrix is 0, the matrix has no inverse;
when the determinant is not 0, the matrix is invertible (non-singular).
45
46
Chapter 3. Determinants
Reading
Study Larson Ch 3.1. Note the determinant of a 1 1 matrix
(the entry itself) in Ed 4, top of p 113 (or Ed 5, top of p 121).
Find the determinant of a 2 2 matrix in Ed 4, p 112 (or
Ed 5, p 120).
Study how these are extended to define the determinant of a
square matrices of any size.
Note that you can expand by any row or column: the answer will be the
same. Of course we choose the easiest one: the row or column with the most
0s.
In the next section we will see how we can simplify rows and columns and/or
create more 0s, to make this task even easier.
Note that the cofactor of any entry has two parts to it:
a plus or minus, following a chequerboard pattern, and
a determinant: that of the matrix left after deleting the row and column containing that entry.
Two warnings on shortcuts:
Do not use the alternative forward and backwards diagonal method (p
117 of Ed 4, or p 125 of Ed 5) on matrices bigger than 3 3. It does
not work because it omits many terms.
The shortcut for triangular matrices (p 117/8 of Ed 4, or p 126/7 of
Ed 5) works for entries on the main diagonal. Be careful if you want to
apply it to a triangular matrix where the entries lie on the backwards
diagonal: matrices of order 6 or more may need a sign adjustment.
Activity
Master Section 3.1 Exercises 1, 3, 9, 11, 17, 21, 23, 31.
Also do Ed 4: 35, 39, 41, 43, 47, 48, 51;
or Ed 5: 37, 43, 45, 47, 53, 54, 57, and 39 with technology.
47
48
Chapter 3. Determinants
Reading
Study Larson, Chapter 3.2, noting Theorem 3.3. Note that
we may use column operations when simplifying a matrix to
find its determinant. We could not do that when solving systems of linear equations because the rows, not the columns,
represented the equations.
Activity
Section 3.2 Exercises, master odd numbers in Ed 4: 1-21, 29,
33, 37, 39 (or Ed 5: 1-21, 31, 35, 39, 41, 43). Try others too.
3.3
Properties of Determinants
Reading
Note the properties of determinants listed in each of the Theorems in Larson 3.3, particularly Theorem 3.7.
Note Example 2: when finding determinants, we draw out
factors from one row or column at a time.
This means that if A has 3 rows, 10A has determinant
10.10.10 det(A) = 1000 det A.
Activity
Master the Section 3.3 Exercises, 1-31, 32, 35.
Also do Ed 4: 37, 47, 48 (or Ed 5: 39, 49, 50).
49
50
Chapter 3. Determinants
Activity
You may find it useful to review the Chapter 3 Summary on
pp 152/3. Grossmans summary on Determinants is also included in the coloured pages in the Study Guide Appendix A.
Note, however, that for this course you are not required to
study eigenvalues, adjoints, and Cramers Rule.
Activity
Try some of the Chapter 3 Review Exercises.
The Cumulative Test covers Chapters 1 to 3.
Chapter
Complex Numbers
We know that there is no real number that squares to give a negative number.
Hence there is no real number that is the square root of -1, or -4, etc.
In this chapter we show how creating new numbers whose squares are negative, and using them to build the so-called Complex Numbers, gives a powerful extension of our number system.
Scientists, mathematicians and engineers use them for a host of real-world
applications: to perform rotations, solve equations, and describe physical
phenomena.
In Algebra & Calculus I, we use complex numbers to solve some polynomial
equations. In Algebra & Calculus II, you will use complex numbers to solve
linear differential equations in Calculus too.
51
52
Activity
Do Appendix B, Section 8.1 Exercises: odds 1-43, 51 and 56.
Do more if you can manage, and think about question 58.
4.2
53
Reading
Study Appendix B, Chapter 8.2, p 439 - 44.
Note also the link with vectors. Some but not all of the geometry of vectors
will extend to complex numbers. We talk about a complex number as having
modulus. This is equivalent to a vector in R2 having length or magnitude.
Recall, in particular, that when you subtract two position vectors, ~a ~b,
you get the vector that joins their endpoints. Hence ||~a ~b|| measures the
distance between their endpoints.
Complex numbers z and w also define two points in the complex plane, so
|z w| measures the distance between them;
eg
|(3 + 2i) (1 + i)| measures the distance between points (3, 2) and (1, 1);
|z 2| measures the distance between points z = a + ib and 2 = 2 + 0i.
We can use these ideas to describe a circle: for example, the equation of the
circle with centre (2, 1) and radius 3 is:
(x 2)2 + (y 1)2 = 9 in cartesian form,
||~r (2, 1)|| = 3 in vector form
|z (2 + i)| = 3 in the complex plane
where ~r is the variable vector (x, y) from the origin and z the variable number
a + ib or point (a, b).
Activity
Do Appendix B, Section 8.2 Exercises, 1, 3, 4, 11, 12, odds
13-23, 27, 31, 33.
The notes above, on the geometric similarity to vectors, may
help you with 31 and 32.
Think about how you would prove the results in 29 and 30.
Do 34 if you wish.
54
Reading
Firstly study Appendix B, p 445 to mid 447. Be very competent at transforming between a + ib form and polar form
r(cos + i sin ). Sketch the point in the correct quadrant
first, to ensure correct choice of angle or signs.
Activity
Do Section 8.3 Exercises, 1-4, odds 5-13, 17, 19, 23.
Reading
Then study mid 447449, carefully. You need the following
identities:
cos (1 + 2 ) = cos 1 cos 2 sin 1 sin 2
sin (1 + 2 ) = sin 1 cos 2 + cos 1 sin 2
Be sure to establish that
when we multiply two complex numbers, their angles add;
when we divide them, their angles subtract;
when we raise a complex number to a positive integer power, the angle
multiplies by that power.
The last result is de Moivres Theorem: obtained by simply applying the
first result repeatedly.
55
Activity
Note that as a result of the conclusions of Theorem 8.4, some
more tedious complex number calculations of multiplication,
division and powers can be done more easily in polar form
than a + ib form. These are even better done in Eulers
exponential form, which we introduce in the next section.
Do a few exercises from Section 8.3, 27-44, 36 and 38. Do
more when you have met Euler form. Gain enough experience
to be able to use your judgment to decide in advance which
method is likely to be easier or quicker.
Reading
Study Appendix B, p 450452 on nth roots of complex numbers. These may be done more intuitively in Euler form, so
re-visit the process and exercises once you have Euler notation.
Note that the nth roots of z have
the same modulus (length): the nth root of |z|;
arguments (angles) that differ by
2
n .
So exactly n of them will fit into one revolution of 2 radians, before they
start to repeat.
To plot them it helps start with the easiest one (found by setting k = 0). It
has argument n where is arg z. Be careful not to plot it at the same angle
as z; nor at the same distance.
Conclude that there are exactly n different nth roots: 3 cube roots, 5 fifth
roots, etc. When plotted, they lie equally spaced by angle 2
n around a circle
1
of radius |z| n .
Activity
Do a few exercises from Appendix B, Section 8.3: 27, 33, 35,
37. These are often more easily done in Euler form, so come
back to them in the next section.
Try 57, 59 and 61, using the fact that the solutions x can be
expressed as nth roots of some complex number.
56
Reading
Refer to Appendix B, p 448, Theorem 8.4.
Notice that the angles 1 and 2 behave like exponents:
when we multiply z1 and z2 , their angles add;
when we divide, they subtract;
when we raise z to a power, its angle is multiplied by that power.
To remind us this is how the angles behave, when doing complex number
algebra, it seems it might be helpful to write the angle as though it were an
exponent: only the angle though, not the modulus r, which behaves quite
ordinarily.
Remind yourself by plotting z and showing a, b, r and on your diagram,
why it is that if z = a + ib has polar form
z = r(cos + i sin ).
Suppose we write just the angle part (cos + i sin ) in exponential notation
instead, as ei : ie
cos + i sin = ei
Then the polar form is simply rei , so we write
z = rei .
We refer to this as the Euler form of the number, or Euler notation.
57
5 5
3
( 2ei 4 )5 (2ei 6 )3 = ( 2 ei 4 ) (23 ei 6 )
3i
= 32 2 e 4
3
3
+ i sin
= 32 2 cos
4
4
1
1
= 32 2 + i
2
2
= 32 32i.
which probably offers
a neater method than having to multiply 5 brackets of
(1 i) by 3 of ( 3 + i).
12
i/4
2e
3 + i = 2ei/6
So
1+i
3+i
12
2ei/4
2ei/6
!12
58
=
=
=
=
1 12 ei3
ei2
2
1 i
e
26
1
(cos + i sin )
64
1
1
(1 + 0i) =
.
64
64
Note that Euler form will clearly help with multiplication, division and
powers, but that it is not useful for addition or subtraction, where form
a + ib is best.
Activity
Now do Appendix B, Exercises 8.3, 27, 33, 35, 37, using Euler
form. Be careful to choose the correct quadrant angle.
Also prove the following important result:
if z = rei then z = rei .
Note on Euler notation:
We write the argument as an exponent, because it behaves like one. But
why do we include i in the exponent, and why use base e?
For a number of reasons: there are a number of areas in which complex
numbers behave like exponentials with base e. Taylor series is one, but
these are done in Algebra & Calculus II.
Let us look at an argument from calculus, instead:
we know that for real b, if we differentiate ebx we get bebx , ie we get the
original function back again, but with coefficient b.
But if we differentiate cos x + i sin x with respect to x, we get
sin x + i cos x = i(cos x + i sin x)
ie the original back again, but with coefficient i.
So the i behaves like b in an expression of form ebx .
So cos x + i sin x behaves like eix when we differentiate it.
59
60
Equating moduli and arguments both sides, because they represent the same
complex number, we get
rn = a and n = b + k2
so solving for r and gives
r = a1/n and =
b + k2
n
where k is any integer, but in fact the angles repeat after k consecutive
values, so there are only k distinct (different) roots.
So the nth roots of aei(b+k2) are
a1/n ei
b+k2
n
Note that it looks as though we take nth roots in the conventional real
number way, but on the generalised angle!
The z-values
are
the
cube
roots
of
the
right
hand
side,
so
we
plot
4
3 + 4i
or 4( 3 + 1i) and find its Euler form, then we generalise the angle:
4 3 + 4i = 8 ei 6 = 8 ei( 6 +2k) .
/6+2k
3
61
Of course if we can find one easy root, we can find the others geometrically:
Example 4.6 Find the cube roots of -8.
We know there is one cube root, the real-valued one, -2. Plot this one on the
negative x-axis: then the other 2 complex cube roots follow geometrically,
spaced evenly around the same circle.
(Alternatively, write -8 in Euler form as 8ei , generalise the angle, and then
take power 1/3, etc.)
Activity
Plot the three cube roots of -8 found above, deduce their
moduli and arguments from your diagram, express each in
Eulers form, rei , convert each one to a + b form, cube them
and check that you do indeed get -8.
If we cannot find an easy root to get started, we simply work algebraically.
Activity
Find the 4th roots of i. It is not easy to guess what they will
be: there is no obvious 4th root. We can work algebraically
rather than geometrically: plot i to see that its modulus
is 1 and its argument is /2. In Euler form, i = 1ei/2 .
Generalise the angle by adding k2, then proceed to take
4th roots. When you have found 4 answers, check them by
raising them to power 4.
Activity
Go back to Appendix B, Section 8.3 Exercises, and do 45,
47, 49, odds 55 - 61 and 64, using Euler notation wherever it
is helpful.
62
b2 4ac
2a
14
1 3
=
2
2
We now view 3 as 3 1 and use the fact that the square roots of -1
are i. We have two complex solutions:
1 i 3
1
i 3
z=
=
.
2
2
2
Notice that the two solutions are a conjugate pair.
Alternatively, we could complete the square, then factorise the difference of
squares that results:
1 3
+
4 4
1 2 3 2
=
z+
i
2
4
!
!
1
1
3
3
=
z+
i
z+ +
i
2
2
2
2
z2 + z + 1 = z2 + z +
63
Activity
How do you know that the equation z 2 + 2z + 2 = 0 has no
solutions in R?
Solve it in C, ie find all complex solutions, by completing
the square or using the quadratic formula. You should find
solutions 1 i. Notice that they are a complex conjugate
pair.
Now let us look at other new techniques we can use with complex numbers:
In R there is no factorisation of a sum of squares, a2 + b2 . Recall
the difference of squares factorisation, however:
a2 b2 = (a + b)(a b).
Now note that in C, because i2 = 1, we can rewrite +1 as i2 . Then
a2 + b2 = a2 i2 b2
making it a difference of squares which we can factorise as (a + ib)(a ib).
So we have a factorisation for a sum of squares in C:
a2 + b2 = (a + bi)(a bi)
Being able to factorise a sum of squares is helpful for solving equations in
C.
Example 4.9 Solve z 2 + 4 = 0 in C, ie for all complex solutions z.
64
z 2 + 2 = z 2 2i2 = (z + i 2)(z i 2)
65
Activity
Factorise z 3 +3z 2 +9z13 using the method suggested above:
show that z = 1 makes this expression 0, and hence (z 1)
is a factor. Then use trial and error or long division to show
that the other factor is z 2 + 4z + 13. Hence show that the
four complex solutions of the equation
z 3 + 3z 2 + 9z 13 = 0
are z = 1 and 2 3i. (Notice the complex conjugate pair
again.)
Example 4.11 Solve z 4 + 4 = 0.
We could write this as z 4 = 4, and notice that z is therefore a fourth root
of -4, and so find the 4th roots using the methods of the last section.
But there are other interesting methods:
We could factorise the sum of squares,
(z 2 2i)(z 2 + 2i) = 0
then solve z 2 = 2i and z 2 = 2i for z by finding the two square roots of 2i
and the two square roots of 2i.
Confirm that these give the same answers as the 4th roots above.
Alternatively, we can complete the square unconventionally, by adding a
middle term. Of course we must subtract it again too.
z 4 + 4 = (z 4 + 4z 2 + 4) 4z 2 .
Now this gives a difference of squares, so it can be factorised:
2
z 4 + 4 = (z 2 + 2) (2z)2
= [(z 2 + 2) 2z][(z 2 + 2) + 2z]
= (z 2 2z + 2)(z 2 + 2z + 2)
Completing the square in each bracket conventionally with the middle term,
gives
z 4 + 4 = (z 2 2z + 1 + 1)(z 2 + 2z + 1 + 1)
= ((z 1)2 i2 )((z + 1)2 i2 )
= (z 1 i)(z 1 + i)(z + 1 i)(z + 1 + i)
which is fully factorised into linear factors at last. So the solutions are
z = 1 + i, 1 i, 1 + i, 1 i.
66
Activity
Solve z 5 + 4z = 0 in C expecting 5 solutions. You may make
use of the results above.
Also solve z 4 4 = 0, by factorisation. Check your answers
by finding the 4th roots of 4.
Activity
Lastly, build a polynomial equation of degree 6 that has solutions 1, 2, 2 + i and 1 i. You may leave it factorised. The
solution is given below, but try it yourself first.
We can reconstruct the polynomial from its factors.
Solutions 1 and 2 must have come from factors (z 1) and (z 2), but we
are given only 4 roots to a 6th degree polynomial equation, so two roots are
missing. Pure real solutions 2 and -1 do not have conjugates which introduce
further solutions, but the other two do. We can find the conjugates of the
67
two complex roots, then express each as a factor instead, then multiply all
6 of the factors to get the full polynomial:
(z 1)(z 2)(z 2 i)(z 2 + i)(z 1 + i)(z 1 i) = 0.
Since any constant multiple is a polynomial of degree 6 with those solutions,
we could offer the following more general form, with k any non-zero constant:
k(z 1)(z 2)(z 2 i)(z 2 + i)(z 1 + i)(z 1 i) = 0.
68
69
eia + eia
,
2
eia eia
.
2i
We know from the above that this holds for any real number a. But we now
extend these formulae to complex numbers too, by simply defining for any
complex number z:
cos z =
sin z =
eiz + eiz
2
iz
e eiz
.
2i
We know that the right hand side has meaning because of the definitions we
have for eiz and eiz above.
We then extend to definitions of the other trig functions of z too, in the
sin z
natural way: tan z is defined as cos
z , and the other 3 trig functions are
defined as the reciprocals of these.
Example 4.13
cos (3 + 4i) = (1/2)(ei(3+4i) + ei(3+4i) )
= (1/2)(e3i4 + e3i+4 )
= (1/2)(e4 e3i + e4 e3i )
= (1/2)(e4 (cos 3 + i sin 3) + e4 (cos 3 i sin 3))
= (1/2)[cos 3(e4 + e4 ) + i sin 3(e4 + e4 )] etc.
The log function:
To define log z we write z in Euler form and generalise the angle, just as we
did for nth roots:
Suppose z = reit = rei(t+2k) , k any integer.
70
Activity
Plot the values of log (16) in the complex plane.
Show that they all have the same real part, ie lie on the same
vertical line, but that there is a vertical step of 2 or roughly
6.28 between their imaginary parts.
Mark the principal value of log (16) on your graph. Why
will the principal value always be the one nearest the x-axis?
Complex Powers
We know what we mean by the power z n when the base z is complex, and
the exponent n is any integer. We have now also defined the exponential
function ez when the exponent is complex but the base is the real number
e.
We now want to define complex powers z w more generally, to cover the case
where the base and exponent are both complex-valued, as in (2 + i)1i or
ii .
71
72
Now
to find log(1 + 3i), we use the Euler form of 1 + 3i. The modulus is
10. For simplicity write t for the argument, ie t = arctan 3.
Activity
Try the above examples yourself: aim at minimal reference.
4.7
73
Try to see the similarities and differences between the topics you have covered. Think about the ways in which complex numbers behave like geometric
vectors in the plane, and the ways in which they do not.
Activity
The Readings do not have a summary of all the aspects of
Complex Numbers that we have covered.
Review the main points, and list them, to reinforce your
learning.
74
Module
II
CLEVER
CALCULUS
75
Overview
The development of calculus took over 2000 years and is arguably one of the
finest achievements of the human mind. The subject had its origins in the
work of Archimedes (c. 287-212 BC) and others, and was further developed
into its more modern form by Newton, Leibniz, and others in the seventeenth
century. Since then calculus has been refined and applied to almost every
area of human endeavor. Anything that represents change can be modelled
with calculus and so you will find it used in engineering, science, business
and even in art and sport.
In this module we begin with a revision of important aspects of functions
and then progress to the foundations of differential and integral calculus.
We cover much of the material from Chapters 1 to 6 of the textbook by
Stewart, but some sections are omitted or covered only a little. As we come
to each section of this module, you will be guided further on how carefully
to study each section of the textbook.
The approach used here is to explore topics and concepts using a range of
geometrical, numerical and algebraic techniques. Consequently there will be
much emphasis in the module on using the computer package Matlab or
the equivalent technology, as an aid to exploring the graphics and numerics
associated with development and understanding of concepts.
The theme in this Module is to represent functions and models using four
techniques: tables; graphs; formulas; and words. That is we use numeric,
geometric, algebraic, and verbal methods wherever possible.
Reading
Background Reading:
Read the Preview of your calculus textbook A Preview of
Calculus.
77
78
Chapter
INTRODUCTION
You will need to work rather quickly through this chapter. Most of the
functions in this chapter will be known to you from high school, although
the section on inverses may be new or unfamiliar. The approach taken to
these functions, and the way they are modelled, will probably be new to
you.
This chapter on functions is very important for your success in developing
calculus. All of the so-called elementary functions (powers, polynomials,
exponential, logarithmic, and trigonometric) to be used in the course are
introduced here.
The important thing for you is to become familiar with graphical and numerical approach to functions which adds so much meaning to common algebraic
methods. The computer package Matlab or its equivalent, is an excellent
aid to this. You need to be able to think more graphically about functions
and to be able to read and interpret tables of numerical data. This will help
you to apply functions in the modelling of real-world phenomena.
There is also a section on constructing new functions from old. How to shift,
flip and stretch graphs.
79
80
Reading
Study the set textbook (Stewart) edition 3, chapter 1 section
1.1, pp 11-21.
Note the use of the terminology domain/range, independent/dependent variable, increasing/decreasing. Also, the interval notation [a, b] and (a, b) are
introduced early in the section. For example: The set of x values such that
1 x < 3 is written as {x|1 x < 3} or simply as [1, 3). Using symmetry
by recognising even and odd functions can sometimes simplify things.
Activity
Do Stewart exercises Sec 1.1 nos 1, 5, 7, 9, 15, 61, 63.
Note that answers are given in the back for many of the odd
numbered questions
Computer Activity
Stewart: section 1.4, no 21.
Matlab note: for example
x=0: 0.01: 5;
f=10*x.^2;
g=x.^3/10;
plot(x,f,x,g)}
Note: for all of these Computer Activities, you are urged to use Matlab
or Scilab. You may of course use any other mathematics package, but we
will only support Matlab and Scilab. Of course you could instead use
a graphics calculator or even a scientific calculator to do these Computer
Activities, but it will take longer and is more tedious.
1.2
81
Reading
Stewart: study section 1.2, pp 25-29.
Activity
Stewart: do sec 1.2 exercises 5,7,11,13.
Computer Activity
A linear supply curve is given by p = 1.05q + 0.02, where
q is the quantity of goods and p is the price of the goods.
The demand curve is p = 5.4 1.1q. Using a computer plot,
find a close approximation to the equilibrium values of p and
q where the supply and demand curves intersect. Matlab
hint: define a q set of points between (say) 0 and 10 using
q=0:0.01:10;. This defines a set from 0 to 10 in steps of 0.01.
Now the two graphs can be plotted one at a time using hold
on to hold the first plot. For example plot(q,1.05*q+0.02)
, and when you have plotted the second graph you can use
zoom and then click on the intersection a few times to read
off an accurate intersection. Check that the intersection is at
about q = 2.5 and p = 2.65.
Polynomials should be quite familiar to you. We normally make use of
polynomials up to about power 4. The most commonly used polynomials
are linear functions (degree 1) and quadratic functions (degree 2).
Reading
Stewart: study sec 1.2, pp 29-31.
82
Activity
Stewart: sec 1.2 exercises 3, 9.
Computer Activity
Stewart: sec 1.2 no 23.
Reading
Stewart: study sec 1.2 pp 31-32.
Activity
Stewart: do sec 1.2 exercises 1(a),(c),3.
Computer Activity
Stewart: do sec 1.4 ex 25.
Rational Functions and Algebraic Functions are now defined.
Reading
Stewart: study sec 1.2, pp 32-33.
Trigonometric Functions. This section should be a revision for you of
high school. Make sure that your calculators are set on radians. When
we write sin x, you must appreciate that x is a number, and so must be
in radians. You should not need to use degrees mode on your calculator
when x is a number. You can only use degrees mode when x represents the
number of degrees of rotation, which is not such a good idea in mathematics,
since mathematics is based on the study of number. Make sure that you
can sketch the sine curve, showing the amplitude and period. The cosine
curve is then just the same curve shifted back a distance of /2. The tan
83
sin
t and
2
sin t = 1 cos t. The inverse trig functions will be seen later.
Reading
Stewart: See Reference inside front cover and sec1.2 pp 3334, and Appendix C on Inverse Trigonometric functions (see
Stewart page A25-A27).
Activity
Use a calculator to find one of the solutions to the equation
2 = 5 sin(3x). (Answer: sin(3x) = 2/5 = 0.4, and so 3x =
sin1 0.4 = 0.4115 (from calculator). Hence, x 0.1372. If
you didnt get that, check if your calculator is set in radians.
Note: in Matlab the inverse sine function sin1 x is written
as asin(x).
Computer Activity
Stewart: sec 1.4 exercise 9.
Exponential, Logarithmic, and Transcendental Functions are briefly
defined and introduced, to be seen in detail later.
Reading
Stewart: study sec 1.2 pp 34,35.
1.3
In this section a summary is given of translations, shifts and stretches, reflections, combinations and composite functions (function of a function). The
composite function (f g)(x) = f (g(x)) is introduced. This is quite often
called a function of a function for obvious reasons. For example, if f (x) = x2
and g(x) = sin x, then (f g)(x) = f (g(x)) = g 2 = sin2 x. Composition of
functions is one of the most useful elements of the study of functions.
84
Reading
Stewart: study sec 1.3 pp 38-45.
Activity
Stewart: do sec 1.3 ex 3, 5, 27, 35, 50.
Reading
Stewart: peruse sec 1.4 pp 49-53. Example 9 on p53 illustrates how a difficult equation can be accurately solved by
graphical means using the zoom facility. In Matlab just
click on the zoom button on the toolbar of the plot and then
click on the portion of the plot which you want to enlarge.
85
Reading
Stewart: study section 1.5 pp 55-61
Take particular note of the general graph of the exponential function ax
exemplified by figure 2 for the case when a = 2. You should be able to
recognise the important features, ie the curve has a horizontal asymptote
for the negative x axis, it then passes through the y axis at y = 1 and then
climbs rapidly.
x note of Figure 4 also. For example, if a = 0.5, then the
Take
1
graph is y = 2 = 2x .
Note that the x (or t) is the independent variable and often denotes position
(or time) for the function ax (or at ). Take note of families of exponential
functions shown in Figure 3, the laws of exponents, and applications. The
number e is very special. Remember that it has a value of about 2.718, and
that it is just the value of a which makes ax have a slope of 1 at the point
(0,1) on the y axis. We will see soon that the function y = ex is the only
function whose slope at any point (x, y) on the curve, is equal to the value
of the function itself. In other words, the rate of change of ex is equal to ex .
It is the only function in the entire universe that has this property.
Activity
Stewart: do sec 1.5 exercises 13,17,19.
Computer Activity
Stewart: do sec 1.5 exercise 25.
1.6
Reading
Stewart: study sec 1.6 pp 63-67.
86
Activity
Stewart: do sec 1.6 exercises 3,15,19,21.
Note that the graph of an inverse function is the mirror image of the function
itself reflected about the line y = x.
Another neat property of inverse functions is that they undo one another
algebraically. Thus f (f 1 (x)) = f 1 (f (x)) = x. For example if f (T ) =
4T 160, then T = f4 + 40, ie the inverse function is f 1 (T ) = T4 + 40.
Then f (f 1 (T )) = 4f 1 (T ) 160 = 4( T4 + 40) 160 = T . Also f 1 (f (T )) =
f (T )
4T 160
+ 40 = T .
4 + 40 =
4
Logarithms are inverse functions for exponential functions. Make sure that
you can sketch the graph of a log function. The easiest way to do that is
to remember how to sketch the exponential function, and then just take the
mirror image about y = x.
Take particular note of the three main laws.
Note that power functions with positive exponents always increase more
rapidly than log functions.
We mainly concentrate on the natural logarithm function to base e and will
write it as ln x. Note that Matlab uses the notation log(x) as the natural
logarithm function.
We wont normally need to change to other logarithm bases, but if you ever
need to do that, remember to use formula 10 on the bottom of page 69 of
Stewart.
Reading
Stewart: study sec 1.6 pp 67-71.
Activity
Stewart: sec 1.6 do exercises 23, 25, 37, 49.
Chapter Summary The last section of this Chapter, on parametric curves,
is to be omitted. It will be studied in the course MAT2100.
This first chapter is largely a revision of the function work you studied in
high school. But the material forms a very important basis for the tools
needed to develop the ideas and the applications of calculus.
Chapter
INTRODUCTION
In this chapter, the derivative is introduced as an instantaneous rate of
change. Initially, tangents and speed are used to motivate practical examples, and then the limit definition is developed. By the end of the chapter
you should be able to:
(i) understand the notion of an average rate of change;
(ii) understand the notion of a limit and be able to find the limiting value
for simple examples;
(iii) find derivatives numerically by taking arbitrarily fine difference quotients;
(iv) visualize derivatives graphically as the slope of the graph when you
zoom in (using your computer);
(v) interpret the meaning of first and second derivatives;
(vi) understand the meaning of an antiderivative.
87
88
Reading
Stewart: study sec 2.1 pp 93-97.
Activity
Stewart: sec 2.1 do exercises 3, 5, 7.
89
In this section we use numerical tables and graphics to see what the value
of limits might be.
Reading
Stewart: study pp 98-105.
Example 4 shows that there are some dangers in just guessing
the value of a limit by using a table of some values near the
limiting point. Figure 7 illustrates the complex behaviour of
the function when you get near x = 0. Strictly, we require
that the left-hand and right-hand limits are the same. Then
we say that the limit of the function exists and is given by
their common values. Examples are given of some weird and
wonderful functions. Note: the precise definition of a limit
is given in Appendix D. It is indeed precise, but is a rather
abstract definition, and is not examinable.
Activity
Stewart: sec 2.2 do exercises 1, 3, 13.
Computer Activity
Stewart: sec 2.2 exercise 23. From your answer to part (a),
look back at sec 1.5. For part (b), plot for values of x from
2 to 3.
Matlab Note for sec 2.2 exercise 23 part (b)
h=0.001; x=-2:h:3; y=(1+x).^(1./x); plot(x,y)
2.3
In this section, we develop methods for finding the exact value of limits. To
do this we basically make use of five common sense Limit Laws, and the
Direct Substitution Property for simpler functions.
90
Reading
Stewart: sec 2.3 study pp 108-113.
If we are lucky, the value of limxa f (x) = f (a). The Direct
Substitution Property is used here. But there are many weird
functions for which it is not so easy. Then we need some
tricks. Some of these tricks are exposed in Examples 1 to
7. Note: dont worry too much about the last part of this
section from the start of Example 9 to the end of the section.
You might read it for interest, but it is not examinable.
Activity
Stewart: sec 2.3 do exercises 1(a,f,g,h), 9.
2.4 Continuity
Continuity of a function at a point can be thought of geometrically as no
sudden and abrupt change of position near or at the point. The shape can
change at the point, but there can be no breaks there. A good example of
this is seen in the function |x| shown in Stewart:Figure 3, section 2.3. The
function is continuous at x = 0, even though there is a dramatic change in
shape there. Numerically, we say that values of the function near or at the
point change gradually.
Reading
Stewart: sec 2.4 study pp 117-122.
Take note of Definition 1 and work through Examples 1 and
2. Note Definition 2 and 3 and work through Example 4.
Theorems 4 and 5 are intuitively to be expected and the
proofs are easy. Study Example 5 and take note of Theorem
7 which you will expect to be true anyway from their graphs.
The rest of the section may be omitted, but should be read
for interest by students majoring in mathematics.
Activity
Stewart: sec 2.4 do exercises 1, 3, 15.
2.5
91
The important thing in this section is that you understand vertical asymptotes where limxa f (x) = and that you understand horizontal asymptotes where limx f (x) = L.
Reading
Stewart: sec 2.5 study pp 128-137.
Make sure you understand that is not a number, but is a
symbol for the idea that the process is getting bigger without
bound. Dont worry about the precise definitions given in the
Appendix D of the textbook. The ten Examples give good
illustration of the concepts. You might omit Example 6.
Activity
Stewart: sec 2.5 do exercises 3, 21(hint: see Example 5),
47(b).
Computer Activity
Stewart: sec 2.5 do exercise 11, 13.
2.6
For tangents, you probably learnt the value of the slope as the rise over
the run. If you are given a curve or line whose equation is y = f (x) and
you take a run, or change in x, of x = h, then the rise is the change
in y given by y = f (x + h) f (x). Then the slope of the tangent is
f (x+h)f (x)
y
approximately rise
and the approximation gets better
run = x =
h
and better as x = h gets smaller.
For moving objects travelling at a constant velocity, the velocity is found by
calculating displacement
. The displacement, or distance from the starting
time
point, is normally given as a function of time, ie f (t). When the objects
velocity is changing, but we wish to calculate its velocity at a particular
time t = a, we have to get smarter. In that case we take a value of t near a,
92
say a + h, and find the average velocity for t between a and a + h. This is
(a)
just f (a+h)f
, ie the distance travelled between t = a and t = a + h. Then
h
we shrink h and find the limit as h 0 giving the instantaneous velocity at
t = a.
Reading
Stewart: sec 2.6 study pp 139-145.
Take particular note of the notation introduced on p143.
is the Greek letter corresponding to our d, and signifies
difference, or change. Note: the ratio
y
f (x + h) f (x)
=
x
h
is called a difference quotient. We will often use this term.
Activity
Stewart: sec 2.6 do exercises 3, 5(a,b), 15, 25.
Computer Activity
Stewart: sec 2.6 do exercise 5(c).
2.7 Derivatives
Rates of change occur in almost all phenomena in our lives and throughout
our immediate environment and the rest of the universe. Given a function
y = f (x), we denote by f 0 (a) the rate of change of f at a particular value of
x = a. We call f 0 (a) the derivative of The function f at a. By the way, the
study of derivatives is called Differential Calculus. So differential calculus
is just about how things change. In Chapter 5 we will start on Integral
Calculus, which is another story.
Reading
Stewart: sec 2.7 study pp 148-152.
Work through Examples 1, 2,3,4. Take particular note of
Examples 5 and 6. Here you can see how to use the difference quotient to help you interpret the units of the derivative
which is an important aid to explaining the actual contextual
meaning of the derivative in different practical situations.
93
Activity
Stewart: sec 2.7 do exercises 1, 7, 29, 31.
Computer Activity
Stewart: sec 2.7 do exercise 11.
2.8
We have seen above how to find the derivative f 0 (a) at a particular point.
If we can do that at every point in an interval, for example 2 x 3, we
will then know the derivative function f 0 (x) over that domain of x values.
Reading
Stewart: sec 2.8 study pp 155-165.
In Example 1, the slope is found at each point of Figure 1
and thus the derivative is known and the sketch of the derivative function follows as shown in Figure 2(b). Example 2
shows how to use difference quotients to calculate good approximations to the value of the derivative at a collection of
points, and thus the derivative function can be sketched. Do
Example 3, dont worry too much about Example 4, and Example 5 is a little harder. Take particular note of alternative
dy
is a
notations for the derivative. The Leibniz notation dx
particularly good one, since it reminds us that
dy
y
= lim
.
dx x0 x
Pages 160-162 are a commentary on differentiability. Essentially, a function f (x) is differentiable on some domain like
2 x 3, if the graphical representation is smooth. Thus
the plot can have no abrupt change of character, since the
rate of change is smooth. Take particular note of higher
derivatives and their notations and interpretations.
Activity
Stewart: sec 2.8 do exercises 1, 19, 37, 39.
94
Matlab Note: given the function y = f (x) = ex the following code will
plot, between x = 0 and x = 1, values of the derivative function f 0 (x) which
2
just happens to be f 0 (x) = 2xex . First of all the function f is plotted, and
then the function f 0 is plotted in red for comparison. Note how Matlab
does comments. Anything in the line after the % is taken as a comment.
Note also the need to use .* and .^ when multiplying or taking powers of
an array of numbers, such as x.
h=0.01; % the step size for x
x=0:h:1; % the set of x values
y=exp(x.^2); % the set of y values
plot(x,y);
hold on; % holds the plot to overlay the next one
deriv=2*x.*exp(x.^2); % the derivative values
plot(x,deriv,r) % plot the derivatives in red
In the above, we have used the known function for the derivative. If we
dont know the derivative function and we want to find approximate values
(x)
for the derivative, we can use the difference quotient f (x+h)f
at a set of
h
x values all at one time using Matlab . The trick is to calculate the values
of f (x) and then to calculate the values of f (x + h) by shifting x to x + h.
Then the difference quotients can all be calculated at the same time and
plotted as follows:
h=0.01; % the step size for x
x1=0:h:1; % the set of x values
y1=exp(x1.^2); % the set of y(x) values
plot(x1,y1);
hold on; % holds the plot to overlay the next one
x2=x1+h; % shifts all the values of x1
y2=exp(x2.^2); % the set of y(x+h) values
deriv=(y2-y1)/h; % the approx derivative values
plot(x1,deriv,r) % plot the derivatives in red
Or, if an even shorter way of coding is used:
h=0.01; % the step size for x
x=0:h:1; % the set of x values
plot(x,exp(x.^2)); % plots the function
hold on;
% now plot approximation to the derivative in red
plot(x,(exp((x+h).^2)-exp(x.^2))/h,r);
95
Computer Activity
Problem: given the function y = f (x) = sin x for 4
x 4, use technology to approximate the derivative using
step size x = h = 0.1, and then plot the function in blue
together with the derivative in red. (See figure following for
the answer.)
Stewart: sec 2.8 do exercise 41.
sin x
derivative
2
1
0
1
2
3
4
2.9
0
x
If we know something about the derivative function f (x), how can we find
out about the function itself f (x)? So we need to be able to reverse our
understanding of the rate of change process. For example, if someone measured the speed of your car at certain times and you then needed to find
out where the car was at those times, you would need to reverse the rate of
change process. Formally we call this antidifferentiation and a knowledge of
how to do this turns out to provide very powerful tools for solving problems.
This section gives us a graphical introduction to this reversing process.
Reading
Stewart: sec 2.9 study pp 168-172.
96
Activity
Stewart: sec 2.9 do exercises 1, 7, 27.
Computer Activity
Stewart: sec 2.9 do exercise 29.
Matlab Note on exercise 29 above.
x=linspace(0,4*pi,2000); % 200 points between 0 and 4pi
deriv=exp(-x/10).*sin(x);
plot(x,deriv);
grid; % puts a grid onto the plot
xlabel(x); % how to do an axis label
title(Now hand plot the function which gives this derivative);
Chapter Summary
By now you should have a good idea of the notion of a derivative from a
graphical, numerical, and algebraic point of view. Limits are at the basis of
this understanding. It is also important for you to be able to use the difference quotient to approximate the value of derivatives and also to interpret
practical meanings of derivatives and to find the corresponding units of the
derivative function.
Chapter
Differentiation Rules
INTRODUCTION
In this chapter some fundamental rules are developed for working with derivatives. If we had to calculate the limit every time we wanted the derivative
of a function, life would be rather boring and inefficient. It was important
for you to know that the derivative y 0 = f 0 (x) is a limit of a difference
quotient, for example:
y
;
lim
x0 x
or the same thing
f (x + h) f (x)
lim
.
h0
h
But now we develop some rules for making things easier. Firstly we determine the rules for finding the derivative of the elementary functions: polynomials, powers, exponential, trigonometric, and logarithmic. Along the
way we develop rules for finding the derivative of sums and differences of
functions, products and quotients of functions. The most important differentiation rule of all is how to find the derivative of a function of a function. For
2
example, you were told in Section 2.8 that the derivative of y = f (x) = ex
2
is y 0 = f 0 (x) = 2xex . To do this we use the chain rule which will be developed in this chapter. Section 3.8 of the textbook is to be omitted. It will
be studied in the next course.
97
98
Reading
Stewart: study sec 3.1 pp 183-190.
d n
x =
You will not be required to be able to prove that dx
n1
nx
, except for special cases as shown. The general proof
uses the Binomial theorem which should be of interest to
mathematics majors.
d
It is of interest to note that the derivative dx
is a linear
operator which means that the Constant Multiple Rule,
the Sum Rule, and the Difference Rule can be combined
into one to give
d
d
d
(C1 f (x) C2 g(x)) = C1 f (x) C2 g(x).
dx
dx
dx
We will see in sec 3.5 that the derivative of an exponential
function of the form ax is (ln a)ax . When a = e, we get the
d x
simplest of all differentiation rules dx
e = ex . On page 189
of the textbook, e is defined in a very formal way as a limit.
You should be able to recall that e 2.718.
Remember that if you have an equation like
ex+1 = 3, to solve for x you
x+1
just take natural logs of each side to get ln e
) = ln 3, and since log is
inverse to exponential, you then get x + 1 = ln 3, x = ln 3 1 0.0986 (from
my Matlab log(3)-1). Then you can solve problems like the following:
Example
Find where the tangent to the curve defined by y = ex + 2x 6 is parallel
to the line y = 4x + 6.
Solution The slope at any point on the curve is y 0 = ex + 2 and the slope
of the line is 4. Hence
ex + 2 = 4, ex = 2, x = ln 2 0.6931. At x = ln 2, y = eln 2 + 2(ln 2)
6 = 2 + 2 ln 2 6 2.6137, and so the tangent, at the point (x, y)
(0.6931, 2.6137) on the curve, has slope 4. See following figure.
99
25
y = ex + 2x 6
y = 4x + 6
Solution Point
20
15
10
5
0
5
0
0.5
1.5
x
2.5
Activity
Stewart: sec 3.1 do exercises 3, 7, 11, 13 (remember that
is a constant), 23, 41, 45, 49.
Computer Activity
Stewart: sec 3.1 exercise 33, 50.
3.2
d
Dont be fooled into using dx
(f (x)g(x)) = f 0 (x)g 0 (x). It is a very common
error. As you can see in this section the rule is
d
(f (x)g(x)) = f 0 (x)g(x) + f (x)g 0 (x).
dx
So in fact you must differentiate each function one at a time and add the
results. This rule extends to products of more than two functions as follows:
d
(u(x)v(x)w(x)) = u0 (x)v(x)w(x) + u(x)v 0 (x)w(x) + u(x)v(x)w0 (x),
dx
and so on. The quotient rule is useful and should also be learned. Dont
forget to divide by g 2 and not just g. I sometimes forget!
100
Reading
Stewart: sec 3.2 study pp 193-198.
A note on notation: In Example 3 on p195, g 0 (4) = 3 means
that the derivative g 0 (x) at x = 4 is 3.
Activity
Stewart: sec 3.2 do exercises 1, 3, 5, 13, 29, 31, 37(c), 40, 41.
Computer Activity
Stewart: sec 3.2 exercise 23.
Reading
Stewart: sec 3.3 study pp 200 to the top of p204.
The key concept in Example 1 is that if you are given the
position s of an object as a function of time t, then the velocity is a measure of change. It is the rate of change of
position with respect to time. The average velocity over a
time period t is s
t and the limit of this ratio as t 0
is the instantaneous velocity v(t) = ds
dt . The acceleration is
rate of change of velocity a(t) = dv
which
can also be written
dt
2
0
00
as v (t) or s (t), or in the Leibniz notation ddt2s . Examples
2 and 3 give important applications to density and electric
current problems. We will certainly see more about these
applications in the following course MAT2100.
Activity
Stewart: sec 3.3 do exercises 1, 7, 13, 15, 17.
3.4
101
The derivatives of sin x and cos x are easy to remember. Check carefully to
see the relationship between the graph of the sine function and its derivative,
the cosine function.
Reading
Stewart: sec 3.4 study pp 213-218.
You will not be expected to formally prove that
sin x
=1
x0 x
lim
Activity
Stewart: sec 3.4 do exercises 5, 7, 9, 31, 37.
Computer Activity
Stewart: sec 3.4 do exercise 19.
3.5
102
Reading
Stewart: sec 3.5 study pp 220-225.
You should take note of the informal proof on page 221, but
you are not required to learn the formal proof on page 227.
Take particular note of the power rule for composite functions
given by formula 4 and the proof that the derivative of the
exponential function ax is ax ln a given by formula 5.
The subsection on tangents to parametric curves is to be
omitted.
Activity
Stewart: sec 3.5 do exercises 1, 3, 5, 17, 19, 23, 41, 43, 47,
59, 79.
Computer Activity
Stewart: sec 3.5 do exercise 63.
103
Reading
Stewart: sec 3.6 study pp 232-238.
Note how important the chain rule is in implicit differentiation. Example 1 on pp 233 and 234 is a good introductory
illustration of this.
Make sure that you can prove the formulas for the derivatives
of the inverse functions sin1 x and tan1 x. These are useful
in calculus.
Activity
Stewart: sec 3.6 do exercises 1, 3, 7, 31, 33, 51.
Computer Activity
Stewart: sec 3.6 do exercise 21
3.7
The most important result in this section is the somewhat surprising fact
d
that dx
ln x = x1 . The easiest way to prove this is as follows:
write y = ln x. Then ey = eln x = x, ie ey = x. Now differentiate implicitly
dy
dy
= 1, (by the chain rule). Hence dx
= e1y = x1 ,
with respect to x to get ey dx
as required.
Reading
Stewart: sec 3.7 study pp 240-244.
Example 6 is interesting. The derivative of ln |x| is x1 and so is
positive when x is positive and negative when x is negative.
This result will be used when we get to integration in the
following chapters.
Take note of logarithmic differentiation and its use in finally
d n
proving the power rule, that dx
x = nxn1 for all real numbers n.
Omit p255.
104
d x
dx x
= xx (1 + ln x).
Activity
Stewart: sec 3.7 do exercises 3, 9, 17, 19, 25, 33.
Chapter Summary
Section 3.8 is to be omitted. Try some of the Review problems on pp 255257.
In this chapter you have developed some important tools for differentiating
functions. The most powerful of these is the chain rule. Familiarity with
these rules is essential for later courses in solving differential equations and
is immediately necessary in this course for the two chapters on integration.
Chapter
Applications of
Differentiation
INTRODUCTION
In this chapter we make use of differentiation rules to develop some more
powerful methods involving the chain rule and some tests and methods for
solving optimisation problems. Some sections will either be omitted or covered very lightly. The last section on antiderivatives is an important lead
into integration.
105
106
dA
dt
Reading
Stewart: sec 4.1 study pp 263-266.
Work through each of Examples 1,2,3. The Strategy given
on page 265 can be very helpful. Note the warning given
about substituting known values before you have the chance
to work the rates of change. Work through Examples 4 and
5.
Activity
Stewart: sec 4.1 do exercises 3, 5, 7, 11(note: mi means
miles), 31*(note: * for harder problems), 37*.
107
Reading
Stewart: sec 4.2 study pp 269-274.
This section introduces some important terminology (ie language) for discussing maxima, minima, and other critical
points. The Closed Interval Method on page 273 gives a
good summary of the way to determine absolute extrema.
Work through the seven Examples.
Activity
Stewart: sec 4.2 do exercises 5, 17, 19, 23, 27, 35, 37, 59*.
4.3
In this section we develop ideas on the shape of curves using the First Derivative Test and then use a knowledge of concavity to develop the Second
Derivative Test for finding Local Maxima and Minima. The Mean Value
Theorem is to be omitted, but should be read for interest by mathematics
majors.
Reading
Stewart: sec 4.3 study pp 280-284.
Increasing/Decreasing functions, The First derivative Test,
Concavity, and the Second Derivative Test, are all important.
Make sure that you know when a curve is concave up (ie
f 00 > 0, or the slope is increasing), and when it is concave
down (ie f 00 < 0, or the slope is decreasing). Study Examples
2,3,4 and 5. You might just read Examples 6, 7, and 8 for
interest.
Activity
Stewart: sec 4.3 do exercises 5, 7, 15, 21, 55*.
Computer Activity
Stewart: sec 4.3 do exercise 47*.
108
Reading
Stewart: sec 4.4 study pp 289-290.
Do Example 1 only.
Computer Activity
Stewart: sec 4.4 do exercise 7.
Reading
Stewart: sec 4.6 study pp 306-311.
First of all six important steps in solving optimising problems
are formulated. You should work through each of Examples
1, 2, 3, 4, and 5.
Activity
Stewart: sec 4.6 do exercises 7, 9, 21, 39, 43.
4.7
109
Omit this section. Unless you want to get rich quick. It will not be examined,
but should be read with interest if you are doing a business major.
4.8
Newtons Method
4.9
Antiderivatives
If I tell you that f 0 (x) = 2x and ask you to find the function f (x), you should
have no trouble. Because you remember the power rule for differentiation,
you will tell me in a flash that f (x) = x2 . Careful! f (x) = x2 + 1 also does
the job. So does f (x) = x2 12345. The general answer is f (x) = x2 + C
where C is an arbitrary constant.
In this section, we continue the process of thinking backwards from the
known derivative. To do that we need a very good knowledge of the differd 2
x = 2x and that
entiation rules. In the above, I had to use the fact that dx
d
C
=
0.
dx
We need to think backwards when we know the rate of change of a process,
but we want to find the actual state of the process. For example if you look
at the speedo on a car, you can see what the velocity is, but you might want
to calculate from it your displacement from home in a given time.
Finding antiderivatives means that you have started integrating.
Reading
Stewart: sec 4.9 study pp 327-332.
Activity
Stewart: sec 4.9 do exercises 1, 9, 15, 23(including finding
the two constants), 27(careful when you get to x1 ), 39, 55*.
110
Computer Activity
Stewart: sec 4.9 do exercise 32.
Chapter Summary
Work through some of the Review problems, see pp 335-341.
In this chapter we have worked through a wide range of applications of the
differentiation rules which were developed in the previous chapter. So you
can already see that differential calculus can be very useful. But the story
continues and the true power of calculus awaits us in the next two chapters
where we introduce Integration.
Chapter
Integrals
INTRODUCTION
In a simple sense, differentiation is about rates of change. The difference
quotient y/x reminds us that the derivative is about ratios of quantities.
In a sense, integration is even simpler. It is about summing things to find
quantities.
In this chapter we develop the idea of a Definite Integral which is used to
sum quantities. We find that the definite integral has a rather amazing
connection with the derivative. The connection is called the Fundamental
Theorem of Calculus. The rest of the chapter is devoted to the development
of standard ways to find exact or approximate values of definite integrals.
111
112
Chapter 5. Integrals
Reading
Stewart: sec 5.1 study pp 343-352.
Note that the area can be approximated by rectangles using
the left endpoint of each interval, or using the right interval.
When you take the limit as n , (ie x 0), you get the
same answer either way. So it doesnt matter whether you
use left or right endpoints, in fact you can use any sample of
points, one in each subinterval, as discussed on page 348. To
keep things simpler, we will only use left or right endpoints,
or sometimes the midpoint. The last subsection on The Distance Problem shows that we can calculate approximations
to the distance travelled in the same way, ie distance = velocity time. If we assume that v(t) is constant in each
subinterval t, we get distance = v(t) t. Then we just
sum all of these values across the domain and then take the
limit as t 0, ie the number of subintervals n . This
will give us the exact distance travelled.
Matlab Note:
Here is some Matlab code to find a good approximation to the area under
the curve y = x2 over the domain 0 x 1, using the left endpoints.
h=0.001; % the subinterval width
x=0:h:1-h; % the left subinterval endpoints (1000 values)
y=x.^2; % the height of each rectangle
area=sum(y*h) % the sum function just adds up each value of y*h
The answer to this is: area = 0.3328, which is already close to the exact
answer 1/3. To get even closer, just take an even smaller value like h =
0.0001. If you need to use right endpoints, just change the second code line
to x=h:h:1; , ie x goes from 0 + h in steps of h to 1.
113
Activity
Stewart: sec 5.1 do exercises 11, 13, 15, 17.
Computer Activity
Stewart: sec 5.1 exercises 6, 7.
5.2
Reading
Stewart: sec 5.2 study pp 354-364.
R
Take note of the definition of the definite integral ab f (x) dx
as the limit of a sum of function values multiplied by the
step length x. Note 1, 2, and 3 are important. Note 4 is
interesting, but we will use equal length values of x. Formulas 4,5,6) on page 357 are interesting, but you will not
be required to use them. Study Examples 1,2,3,4 carefully.
Note that the definite integral does not necessarily give the
area under the curve. It is positive when the curve is above
the x-axis and negative when the curve is below. The eight
Properties of the Integral are important. By the way, Properties 2,3,4 indicate that the integral is a linear operator, ie
Z b
a
Z b
a
f (x) dx + C2
Z b
a
g(x) dx.
114
Chapter 5. Integrals
Activity
Stewart: sec 5.2 do exercises 1, 7, 11, 17, 27, 33(draw the
graph), 39, 41, 43.
Computer Activity
Stewart: sec 5.2 do exercise 9 (use technology and increase
n to something big to get more accuracy).
Note on the above Computer Activity:
x3 + 1 in Matlab is sqrt(x.^3+1).
ds
dt ,
Z b
a
so we have
s0 (t) dt = s(b) s(a),
ie, the total change in position is the integral of its derivative. This illustrates the remarkable connection between integration and differentiation.
In a sense, integration and differentiation are inverse operators, since one
unravels the other.
115
Reading
Stewart: sec 5.3 study pp 366-373.
The Evaluation Theorem formalises the discussion above
(where s(t) is in fact the antiderivative of s0 (t)). You can
omit the proof of the Evaluation Theorem on page 367. Take
particular note of Examples
1,2. A good notation for the anR
tiderivative is F (x) = f (x) dx, and is called the indefinite
integral. It is called indefinite, since it is a function, whereas
the definite integral is a single number. The Table of Indefinite Integrals on page 369 should be familiar to you, since if
you differentiate the right hand side, you get the function on
the left. This Table becomes your working set of antiderivatives. Examples 3,4,5,6 are an excellent introduction to using
the Table. The last subsection on Applications gives a very
good indicator of various contexts for the integral.
Activity
Stewart: sec 5.3 do exercises 1, 3, 7, 9, 15, 23, 27, 37, 41, 47,
53, 57.
Computer Activity
Stewart: sec 5.3 do exercises 29, 31, 35.
5.4
In this section the two parts of the Fundamental Theorem of Calculus are
formulated. A good way to remember each part is to think:
1.
2.
d
dx
f (x) dx = f (x);
f 0 (x) dx = f (x).
Reading
Stewart: sec 5.4 study pp 377-382.
You are not required to learn the proof of the first part of
the fundamental theorem, given in the last subsection.
116
Chapter 5. Integrals
et dt = F (x) F (0),
and so
F (x) =
Z x
0
et dt + 3.
Activity
Stewart: sec 5.4 do exercises 5, 7, 21.
Computer Activity
Do the following problem:
A car moves at velocity v(t) = sin(t2 ) + tet . It starts at position s(0) = 0. Use technology to find a good approximation
to its position s(3) at t = 3. (Hint: you wont be able to find
the antiderivative function for sin(t2 ) + tet , so you will need
to look at the Matlab example above for guidance.)
f (g(x))g 0 (x) dx
117
by means of substitution. You will note that integrals of this form have
a complicated function f (g(x)), but that they also have the luxury of the
accompanying function g 0 (x) as a partner. These integrals occur very frequently. You will remember from the chain rule that this type of combination of functions always result when you differentiate a composite function
f (g(x)). So the method of substitution is really a process for unravelling
the chain rule.
Reading
Stewart: sec 5.5 study pp 386-392.
Take note of the different types of substitution that can be
used. It is not always obvious how to choose the substitution,
so make sure that you work each of the Examples 1 to 8.
The last subsection on Symmetry presents ways to sometimes
speed up your calculations.
Activity
Stewart: sec 5.5 do exercises 1, 3, 5, 7, 13, 17, 27 (hint: write
as sec2 x sec x tan x), 33, 41, 61.
5.6
Integration by Parts
uv 0 dx = uv
or,
u0 v dx
Z
u dv = uv
v du.
Reading
Stewart: sec 5.6 study pp 393-398.
Work Examples 1 and 2. Example 3 shows how you sometimes need to apply the integration by parts formula twice.
Example 4 shows that you can sometimes use the cyclic behaviour for trigonometric functions to get there. Take note
of Example 5 which finds the value of the integral and you
can omit Example 6.
118
Chapter 5. Integrals
Activity
Stewart: sec 5.6 do exercises 1, 3, 7, 13, 15.
Computer Activity
Stewart: sec 5.6 do exercise 41 for the case t = 5 and check
your answer by calculating a good approximation with Matlab .
Matlab Note: In the above Computer Activity use v=t.^2.*exp(-t) to
get values of the function. See earlier Matlab code for setting up the left
endpoints and calculating the sum which gives the approximation to the
definite integral.
Reading
Stewart: sec 5.7 study pp 400-402.
Omit the last subsection on Partial Fractions.
Example 1 is a good illustration of setting things up for
a substitution. Example 2 is interesting if you know that
sin2 x = 0.5(1 cos 2x). In this course you dont need to
learn that
Z
119
Activity
Stewart:
sec 5.7 do exercises 1 (hint:
write as
2
2
2
2
sin x sin x cos x and then use sin x = 1 cos x), 7, 29
(hint: substitute u = x0.5 ).
5.8
We will only have a quick look at this section. We do not use Computer
Algebra Systems in this course, so you can omit that subsection.
Reading
Stewart: sec 5.8 study pp 405-407.
Omit the second half of page 407.
Work through Examples 2,3,4. Note that the table of integrals is on References pp 6-10 at the back of the book.
The rest of the section can be omitted.
Activity
Stewart: sec 5.8 do exercises 3, 5, 7.
5.9
Approximate Integration
Reading
Stewart: sec 5.9 study page 412 only.
Here is a reminder of the Matlab code for approximating
R 3 x2
0 e dx:
120
Chapter 5. Integrals
Activity
Stewart: sec 5.9 do exercise 5(a). Note: if you want 8 midpoints, you will need to firstly calculate the value of h. Here
it will be h = 0
8 . In Matlab that is h=(pi-0)/8.
Chapter Summary
The last section on Improper Integrals may be omitted, but should be read
by mathematics majors.
Standard methods of integration have been introduced in this chapter. There
is much useful material here. You are now equipped with the foundation
tools of both differential and integral calculus. This will form the basis of
your abilities as a technician of change. If it moves, you can now hit it with
calculus. The next, and final chapter in this course is the chance to apply
some of these skills to more practical situations.
Before moving on, try some of the Review questions at the end of this
chapter.
Chapter
Applications of
Integration
INTRODUCTION
In this, the final chapter, we study applications of the definite integral including finding areas, volumes, averages, as well as some applications to
physics and engineering. We omit some sections and subsections as described following.
121
122
Reading
Stewart: sec 6.1 study pp 441-445.
Omit the last subsection on Areas Enclosed by Parametric
Curves.
Activity
Stewart: sec 6.1 do exercises 1, 7, 9, 25.
Computer Activity
Stewart: sec 6.1 do exercise 17.
6.2 Volumes
In this section we use Riemann sums to find volumes of revolution as well
as volumes whose cross-sections are known.
Reading
Stewart: sec 6.1 study pp 447-456, but omit the last subsection on Cylindrical Shells.
For volumes of revolution, we will only use the disc method.
The trick here is to find the volume of a disc section of thickness x and radius given by y = f (x). The volume of the
disc is then y 2 x = f (x)2 x. Then sum all the discs to
P
get
f (xi )2 x. This is just a Riemann sumR and in the
limit as x 0 we get the definite integral ab f (x)2 dx
which can be evaluated (or approximated) by the methods
of the last chapter.
Activity
Stewart: sec 6.1 do exercises 1, 3, 5, 11*.
6.3
123
Arc Length
Omit this section which will be studied in the following course MAT2100.
6.4
1
ba
Z b
a
f (x) dx.
Reading
Stewart: sec 6.4 study pp 467-469. You may omit the Mean
Value Theorem for Integrals and omit Example 2.
Activity
Stewart: sec 6.4 do exercises 1, 5, 15.
Computer Activity
Stewart: sec 6.4 do exercise 7.
6.5
Reading
Stewart: sec 6.5 study pp 471-479.
Study the three application subsections and work through
Examples 1-7. Further notes follow.
124
Non-uniform density
In the analysis of this section it has been assumed that the density of the
material or the medium is constant. How do we handle things if the density
is not constant? Here is an example which follows the modelling procedure,
develops the integral, and then we solve it.
Example
Consider a rod from x = 0 to x = 4 whose density at a point x is given by
1
(x) = 2 + x+1
kilograms per metre. Find the mass of the rod.
Solution:
Split the rod up into n subintervals of length x. On a typical subinterval (ie an element of length x) we then assume the density to have
its value at a point xi on the element (eg the left endpoint) and thus be
constant across the element with value (xi ). The mass of the element
is then just (density)(length), or (xi )x. Summing all of the elements
P
gives (xi )x.R This is a left endpoint Riemann sum, and in the limit as
x 0, we get 04 (x) dx.
Thus the mass of the rod is
Z 4
0
R4
1
2+
x+1
2+
1
x+1
dx. Evaluating
Activity
Stewart: sec 6.5 do exercises 1, 15, 23, 35.
Chapter Summary The remainder of this chapter is to be omitted, but
if you are majoring in business, biology, or mathematics, you should read
selections of Sections 6.6 or 6.7.
We have covered a lot of ground in this course and now it is time to study for
the exam. Good luck and hope to see you in the following course MAT2100.
III
APPENDICES A,
B and C:
Readings for
Vectors, Chapter
0
125
127
Appendix A: Vectors
Appendix A contains readings on Vectors for Chapter 0 of Module
1 in this Study Guide.
The readings are an excerpt from Elementary Linear Algebra: International Edition, by Grossman, published by Saunders College
Publishing, pp 154189, 192207. Solutions to odd-numbered exercises are offered in A3337 and A39.