Multivariable Mathematics with Maple

Linear Algebra, Vector Calculus

and Differential Equations

by James A. Carlson and Jennifer M. Johnson

c
1996
Prentice-Hall

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1. Introduction to Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1. A Quick Tour of the Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2. Algebra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3. Graphing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4. Solving Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5. Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
6. Calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
7. Vector and Matrix Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
8. Programming in Maple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
9. Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2. Lines and Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Lines in the Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Lines in 3-space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Planes in 3-space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. More about Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

36
36
39
41
43

3. Applications of Linear Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Temperature at Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Curve-Fitting Polynomial Interpolation . . . . . . . . . . . . . . . . . . . .
4. Linear Versus Polynomial Interpolation . . . . . . . . . . . . . . . . . . . . . . .
5. Cubic Splines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

49
49
52
58
61
64

4. Bases and Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Coordinates in the Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Higher Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. The Vector Space of Piecewise Linear Functions . . . . . . . . . . . . . .
4. Periodic PL Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Temperature at Equilibrium Revisited . . . . . . . . . . . . . . . . . . . . . . . .

67
67
71
74
77
82

5. Affine Transformations in the Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Transforming a Square . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Transforming Parallelograms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Iterated Mappings Making Movies with Maple . . . . . . . . . . . . .
5. Stretches, Rotations, and Shears . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. Appendix: Maple Code for iter and film . . . . . . . . . . . . . . . . . . . .

86
87
89
91
93
95
99

6. Eigenvalues and Eigenvectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Diagonal matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Nondiagonal Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Algebraic Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Diagonalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Ellipses and Their Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. Numerical Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

101
101
102
104
109
113
118

7. Least Squares Fitting a Curve to Data . . . . . . . . . . . . . . . . . . . . . . . .

1. A Formula for the Line of Best Fit . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Solving Inconsistent Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. The Stats Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

124
125
132
134

8. Fourier Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Periodic Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Computing Fourier Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Approximations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. Appendix: Almost Periodic Functions . . . . . . . . . . . . . . . . . . . . . . .

137
137
143
147
149
150
151

9. Curves and Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Curves in the Plane Maps from R to R2 . . . . . . . . . . . . . . . . . .
2. Curves in R3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Parametrizing Surfaces of Revolution . . . . . . . . . . . . . . . . . . . . . . . .

156
156
160
160
162

10. Limits, Continuity, and Differentiability . . . . . . . . . . . . . . . . . . . . . . . .

1. Limits Functions from R to R . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Limits Functions from R2 to R . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Tangent Planes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Differentiability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

168
168
171
172
174
176

11. Optimizing Functions of Several Variables . . . . . . . . . . . . . . . . . . . . . .

1. Review of the One-Variable Case . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Critical Points and the Gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Finding the Critical Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Quadratic Functions and their Perturbations . . . . . . . . . . . . . . . .
5. Taylors Theorem in Two Variables . . . . . . . . . . . . . . . . . . . . . . . . .

181
181
184
184
186
190

ii

6. Completing the Square . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

7. Constrained Extrema . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
12. Transformations and their Jacobians . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Transforming the Coordinate Grid . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Area of Transformed Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Differentiable Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Polar Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. The Area Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. The Change-of-Variables Theorem . . . . . . . . . . . . . . . . . . . . . . . . . .
7. Appendix: Affine Approximations . . . . . . . . . . . . . . . . . . . . . . . . . . .
8. Appendix: Gridtransform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

201
202
205
207
210
212
214
216
217

13. Solving Equations Numerically . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. The Bisection Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Newtons Method for Functions of One Variable . . . . . . . . . . . . .
4. Newtons Method for Solving Systems . . . . . . . . . . . . . . . . . . . . . . .
5. A Bisection Method for Systems of Equations . . . . . . . . . . . . . . .
6. Winding Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

219
219
220
222
224
228
229

14. First-order Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Analytic Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Line Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Drawing Line Fields and Solutions with Maple . . . . . . . . . . . . . .

235
235
239
243

15. Second-order Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. The Physical Basis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Free Oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Damped Oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Overdamping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Critical Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. Forced Oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7. Resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

246
247
247
251
253
254
255
258

16. Numerical Methods for Differential Equations . . . . . . . . . . . . . . . . . .

1. Estimating e with Eulers Method . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Eulers Method for General First-order Equations . . . . . . . . . . .
3. Improvements to Eulers Method . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Systems of Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

261
261
265
268
270

iii

17. Systems of Linear Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . 276

1. Normal Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
2. Direction Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
3. Complex Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
4. Systems of Second-order Equations . . . . . . . . . . . . . . . . . . . . . . . . . . 288

iv

Introduction

The aim of this book, intended as a companion to a traditional text, is to

explore the notions of multivariable calculus using a computer as a tool to help
with computations and with visualization of graphs, transformations, etc. The
software tool we have chosen is Maple; one could as easily have chosen Mathematica or Matlab. In some cases the computer is merely a convenience which
slightly speeds up the work and allows one to accurately treat more examples.
In others it is an essential tool since the necessary computations would take
many minutes, if not hours or days. We will, for example, use Maple to study
the temperature distribution in a thin flat plate by reducing the problem to the
solution of a system of, say, one hundred equations in one hundred unknowns,
then using the resulting numerical data to construct a contour plot which shows
lines of equal temperature. All this could be done by hand, but it would be
laborious work indeed. Such problems would be out of reach without tools for
computation and visualization.
Difficult computations and fancy pictures are, of course, not ends in themselves. We must understand the underlying mathematics if we are to know which
computations to do and which pictures to draw. Likewise we must develop our
own intellectual tools sufficiently well in order to understand, interpret, and
make use of the data and images that we compute. Thus our focus will always
be on the mathematical ideas and their applications. The role of Maple is to
more vividly illustrate them and to widen the range of problems that we can
successfully solve.
To get the most from this book, the reader should work through the examples and exercises as they occur. For example, when the text mentions the
snippet of Maple code
> plot( cos(x) - (1/3)*cos(3x), x = -2*Pi..2*Pi );
the reader should try it out at his or her machine. This particular bit of Maple
will plot the graph of y = cos x (1/3) cos 3x on the interval 2 x 2.
Most chapters can be read independently of the others. However, it is best to
first work through a good part of Chapter One. It is a brief guide to the essentials

Introduction

of Maple: how to do algebraic computations, solve simple equations, compute

derivatives and integrals, and make graphs. It also contains an introduction to
programming in Maple, e.g., how to do repetitive computations using loops, and
how to define new functions.
The great majority of the problems in the text can be solved with just a few
lines of Maple, like the one above for plotting a graph. Occasionally, however, a
paragraph or two of code is required. As an alternative to typing these in, we
have made them available from the web pages at
http://www.math.utah/books/calc2-maple/
You are free to copy any of the code you find there.
As you work with Maple you will sometimes find that things dont work
as you expect. The usual cause of this misbehavior is that computers, unlaik
humans, canit unstond stautements mud with less than purfict spelling, punctuation, grammar and logic. If Maple does not respond or responds with nonsense,
carefully check your work. If it is an example in the book, compare what you
have typed with what is written. Take special care with the placement of punctuation marks like colons and semicolons and the three kinds of quotation marks
single , double ", and backquote . If this fails, take a look at the troubleshooting section at the end of Chapter One, or consult someone with a bit
more experience.
It is always important to think about whether the results of a computation
make sense. Errors in your logic or quirks in Maples thought processes may give
wrong or incomplete answers. The best way to avoid such pitfalls is, as always,
to understand what you are doing.
The authors would like to thank the members of the Calculus II classes
they have taught at the University of Utah for the past three years during the
preparation of this book, particularly Susan Pollock. The Mathematics Department has been generous in its support, and we are grateful to faculty members
Mladen Bestvina, Gerald Davey, Les Glaser, Grant Gustafson, Janos Koll
ar,
Nick Korevaar, Domingo Toledo, and Andrejs Treibergs for their suggestions
and assistance. Special thanks are due Drs. Nelson Beebe, Paul Burchard, and
Michael Spivak for their help at crucial points with the TeX macros.

Chapter 1
Introduction to Maple

The purpose of this first chapter is to give a rapid overview of how one
can use Maple to do algebra, plot graphs, solve equations, etc. Maple can also
compute derivatives and integrals, solve differential equations, and manipulate
vectors and matrices. Much can be done with one-line computations. For example,
> expand((a + b)^3);
expands (a + b)3 to a3 + 3a2 b + 3ab2 + b3 , while
> plot( cos(x) + cos(2*x) + cos(3*x), x = -Pi..Pi );
constructs the graph of the function f (x) = cos x+cos 2x+cos 3x on the interval
[, ], and
> solve( x^2 + 2*x - 5 = 0 );
solves the quadratic equation x2 + 2x 5 = 0.
The best way to learn Maple is by using it. Begin by trying the three
examples above. The symbol > is the prompt, which Maple displays to signal
you that it awaits your command. Commands normally end with a semicolon.
Computers are much fussier about rules of punctuation, grammar, and
spelling than are humans. If something is not working right, check to see if
you are following the rules. For example, Maple will get confused if you say
solve( x^2 + 2x - 5 = 0 ) instead of solve( x^2 + 2*x - 5 = 0 ). Check
for things like misspelled names, extra or missing parentheses. If further thought
doesnt clear things up, ask a human for help. You will soon become an expert
troubleshooter.
Technical details on how to open the Maple program and how to save or
print a Maple file depend heavily on your local system. Thus no information is

Introduction to Maple

given here on these important aspects of getting started. Consult your manual
or local support staff if you need help.
While learning Maple, you will often have questions about how a particular
command or function is used. Fortunately, Maple can help you. To ask about
a command whose name you know, just type a question mark, followed by the
name of the command. Thus
> ?solve
gives information on the solve command. You will probably find the examples
at the end more useful than the technical information at the beginning of the
help file. Here are other things to try:
> ?
> ?intro

1.1

A Quick Tour of the Basics

Below is a sample Maple session, in which we do some simple arithmetic:

> 2 + 2;
> quit
In this session you computed 2+2 by typing a one-line command next to Maples
command prompt >. This is where you type what you want Maple to compute.
You then typed your Return (Unix system) or Enter (Mac version) to tell Maple
to execute your command. Then you typed the command to quit. Alternatively,
just select quit from the file menu.
Maple commands must be properly punctuated: they usually end with a
semicolon. If you forget to type the semicolon, just put it on the next line:
> 2 + 2
> ;
has precisely the same effect as
> 2 + 2;
Addition, subtraction, etc. are standard, and parentheses are used in the
usual way. An asterisk * indicates multiplication and a caret ^ is used for powers:
> ( 1 + 2 ) * ( 6 + 7 ) - 12 / 7;
> 3^(2.1);
Whenever possible, Maple tries to compute exact quantities. Our first command
gives its answer as a fraction, rather than as a decimal, contrary to what we

1.1

A Quick Tour of the Basics

might expect. The second command gives a decimal or floating point answer
because we used floating point forms in our question.
To force Maple to give results in floating point form, use evalf:
> Pi; evalf( " );
> evalf( Pi );
> evalf( exp(1) );
The quote sign " or ditto stands for the most recently computed quantity.
Be aware that Maple distinguishes upper-case letters from lower-case. Thus
evalf(pi) is not the same as evalf(Pi). The function evalf can take a second
(optional) input which determines the precision of the output. (Inputs, called
independent variables in mathematics, are known as arguments in computer
jargon.)
> evalf( Pi, 100 );
For the most part, spacing is unimportant in Maple. In the commands
above, spaces could be omitted without causing any problems. However, thoughtful use of spacing makes Maple code easier to read, and so easier to understand
and, if necessary, to correct.
Standard mathematical functions can be used in Maple so long as we know
their names. To compute the quantities

2 + 14 8 + | 14| sin(1)

and

esin(1.6) +

2 + tan1 (3) :

use
> evalf( 2^(1/3) + 14*8 + abs(-14) - sin(1) );
> exp( sin(1.6*Pi) ) + sqrt(2) + arctan(3) ;
> evalf( " );
This example illustrates another important point. The correct form of a Maple
expression can often be found by intelligent guessing. Thus tan(1) does indeed
compute the tangent, and 20! computes a factorial. If your first guess does not
work, use ? together with your guess to get more information. For instance,
> arctangent( 1.0 );
produces only an echo, but
> ?arctangent
leads to the desired command and examples of its use. The query ?library
gives a listing of all the functions available.

Introduction to Maple

In addition to the functions in the Maple library, there are specialized packages of functions that can be read into the working memory as needed, using
the command with. Try
> ?packages
to see what is available. For example, to work with matrices and vectors, you
must load the linear algebra package. Do this by typing
> with(linalg);
This command produces a list of all the functions in this package and gives you
access to them in your current Maple session. If you close Maple and reopen it
later, you must reload any special packages you want to use.
Once you become familiar with a package, you will prefer to load it using a
colon
> with(linalg):
rather than a semicolon. Again, this gives you access to the linear algebra
commands, but without listing all their names. In general, end Maple commands
with a colon to prevent printing the results on screen. Other packages of interest
are
> ?plots
> ?DEtools
> ?student

1.2

Algebra

Maple knows about variables and algebra. Consider, for example, the expression (a + b)2 . The commands
> ( a + b )^2;
> expand( " );
give the expanded form, and
> factor( " );
brings us back to our starting point. To make long computations easier and
more intelligible, we can assign values to variables:
> p := ( a + b )^2;

b := 1;

p;

1.2

Algebra

In these examples, variables store an expression or a number. Variables can

store almost anything, e.g., a list of points, an equation, a set, a piece of text,
or a function definition:
>
>
>
>
>
>
>

pts := [ [1,2], [3,4] ];

eqn := 2*x - 3*y = 5;
eqns := { 2*x - 3*y = 5, 5*x - 3*y = 1 };
tag := The nth partial sum is;
# backquotes
print( pts, eqn, eqns, tag );
# check
f := x -> x^2;
# define a function
f(2); f(3);
# check definition

Anything we can define or compute in Maple can be assigned to a variable for

future reference. The special symbol := is read gets. Remember that it is
different from the symbol =, used to test equality. Confusing these two symbols
can lead to hard-to-spot errors in your Maple code. Observe that no space is
allowed between the : and the = in an assignment statement.
The symbol # marks a comment. Maple ignores them, but they are useful for
humans trying to read Maple code. On some machines, Maple allows comments
in a multi-line command:
> f := x -> x^2;
f(2); f(3);

# define a function
# check definition

but on other machines Maple ignores everything that appears between the first
comment symbol # and the next command prompt >. On those machines Maple
will not see the second line of commands f(2) and f(3).
Variables can be returned to their original symbolic (unassigned) state. The
commands
> b := b;
> p;
first remove the value 1 assigned to the variable b above and then display the
updated value of p. Similarly,
> unassign(pts, eqn, eqns, tag);
> print( pts, eqn, eqns, tag );
# check
clears the variables assigned above. A more drastic way of clearing Maples
memory is to say
> restart;
This command clears all variables and unloads all packages. Thus, if you need
one later, you must reload it using with.

Introduction to Maple

Pay special attention to the kind of quotes used in examples. The possibilities are the single quote , the backquote , and the double quote ". They all
play different roles.
Here is an extended example of how to use variables and assignment statements.
>
>
>
>
>
>
>

F :=
m :=
F;
a :=
F;
a :=
F;

m*a;
2.1; a := 5;
21.9;
a;

#
#
#
#
#
#
#

Newtons formula for force

set the mass and acceleration
compute the force
reset the acceleration
recompute force
clear a
recompute F

The subs command lets us make temporary substitutions in an expression

as opposed to assigning values. For example, try this:
> g := (a + 1)^2 / ( b - 1)^3 + a / (b - 1);
> simplify( g );
> subs(a = 3, b = 2, g);
Or this:
> subs( a = x + y, b = x + 1, g );
> simplify( " );
a;
b;
Notice that the variables a and b were not permanently assigned a value.
Warning: As you work through these examples and exercises, you may get
strange results because you forgot that you assigned a value to a variable. For
example, if you work Exercise 4 below, you will probably assign the value 5 to
the variable r. If you use r later in your Maple session to mean something else,
say in a problem about polar coordinates, you must first clear it with r := r.
Recall that you can also use the restart command.
Exercise 1. Compute (5/3)21 as an exact expression, as a decimal accurate to
standard Maple precision, and as a decimal accurate to 20 digits precision.
Exercise 2. Compute (x + 1)(x + 2)(x + 3)(x + 4) in expanded form.

Exercise 3. Find the coefficient of a2 b7 in (2a + 3b)9 .

Exercise 4. Use Maple to define the formula for compound interest

A = P (1 + r/100)n ,
where P is the principal, r is the annual interest rate expressed as a percentage,
n is the number of years the principal is invested, and A is the amount of capital
(principal plus accumulated interest) at the end of n years. Set P to $100, r to
5, and compute A. Then compute A with n = 1, n = 5, n = 10. (Your results
will be more understandable in decimal form.)

1.3

Graphing

Exercise 5. Maple has a built-in function for computing partial fraction decompositions. Start with ?partialfractions and follow the clues Maple gives
to find out how to use this function. Use it to rewrite
x2 2x + 5
x4 + 8x3 10x2 104x + 105
as a sum of simpler fractions.

1.3

Graphing

Maple can construct many kinds of graphs, a feature we will frequently use
to visualize mathematical objects and processes. The command
> plot( sin( 3*x ), x = -Pi..Pi );
produces a plot window containing the curve y = sin 3x for x in the interval
from to . No space is allowed between the dots in a plot command. Notice
that the scale on the x-axis is not the same as on the y-axis. To fix this we say
> plot( sin( 3*x ), x = -Pi..Pi, scaling = constrained );
Sometimes it is useful to restrict the range over which y varies. We get a misleading graph from
> plot( tan(x), x = -5..5 );
It does not accurately represent the vertical asymptotes of y = tan x. Better
results are obtained with
> plot( tan(x), x = -5..5, y = -5..5 );
As always: Think about your results. Are they reasonable?
We can plot several curves at once. To plot y = sin x and y = sin 3x together
we say
> plot( { sin(x), sin(3*x) }, x = -Pi..Pi );
Now the first argument of plot is a set of expressions to be graphed. Sets are
enclosed in curly braces and individual items are separated by commas. This
graph can be used to solve the equation sin x = sin 3x. Click with the mouse on
a point where these graphs cross. Maple gives the coordinates of the intersection
point and the x-value is a solution to the equation.

10

Introduction to Maple

Parametric Equations
A curve inthe plane can be described as the graph of a function, as in the
graph of y = 12 4 x2 for x [1, 1], or it can be given parametrically as in
(x(t), y(t)) = (2 cos t, sin t)

for t [0, ].

Often we interpret such a curve as the path traced by a moving particle in the
plane, with (x(t), y(t)) denoting the position of the particle at time t. Use the
plot command to draw a curve from this parametric description:
> plot( [ 2*cos(t), sin(t), t = 0..Pi ] );
The resulting graph is somewhat distorted, because Maple did not use the same
vertical and horizontal scales. There are two ways to fix this. Use
> plot( [ 2*cos(t), sin(t), t = 0..Pi ] , -2..2, -2..2 );
or use the option scaling = constrained.
Polar Coordinates
Polar plots are a special kind of parametric plot. The polar coordinates
(r, ) of points on a curve can be given as a function of some parameter t. In
many cases the parameter is just the angle . Consider the ellipse defined in
standard coordinates by x2 + 4y 2 = 4. To find an equation relating the polar
coordinates r and of a typical point on this ellipse, we make the substitution
x = r cos t and y = r sin t, where t = .
> subs( x = r*cos(t), y = r*sin(t), x^2 + 4*y^2 = 4 );
simplify( " );
solve( ", r );
We find that the ellipse is the collection of points whose polar coordinates (r, )
satisfy
4
r2 =
4 3 cos2
and the following commands draw the right half of the ellipse:
> r := 2/sqrt( 4 - 3*cos(t)^2 );
> plot( [ r, t, t = -Pi/2..Pi/2 ], coords = polar );
Try the examples below. However, before asking Maple to do the plot, try
to predict what the result will be.
> plot( [ 1, t, t = 0..2*Pi], coords = polar );
> plot( [ t, t, t = 0..2*Pi], coords = polar );
> plot( [ sin(4*t), t, t = 0..2*Pi ], coords = polar );
As usual, a more realistic picture is obtained with scaling = constrained.

1.3

Graphing

Plotting Data
Maple can plot data consisting of pairs of x and y values. For example, if
we say
> data := [ [0, 0.53], [1, 1.14], [2, 1.84], [3, 4.12] ];
then data refers to a sequence of five points, (x, y) = (0, 0.53), etc. The result is a
list: something enclosed in square brackets, with elements separated by commas.
Lists are used for collections of objects where the order matters. Individual items
are accessed this way:
> data[1]; data[2]; data[3]; data[4];
In our case, the list items are themselves lists very short ones made up of the
coordinates of a point. To access the second coordinate of the third data point
we say
> data[3][2];
To plot the points in our list we use commands like
> plot( data, style = point );
> plot( data, style = point, symbol = diamond );
> plot( data, style = line, view = [ 0..4, 0..5 ] );
You could also try
> plot( data, style = line, title = Experiment 1 );
Here the backquote is used to specify a plot title. See ?plot[options] for
more information, e.g., about symbols and line styles available. See ?readdata
or ?stats to find out how to read a file of data points into a Maple session.
(You could discover these commands on your own by typing ?data or ?reading
data.)
Exercise 1. Graph y =

x3

6x2

1
on the interval 1 x 4.
+ 11x 6

1
Exercise 2. Graph y =
on the interval 27 x 27. Do your results
3
x
make sense?

Exercise 3. Graph y = tan x and y = x together on 5 x 5. Use the

mouse and your graph to find solutions to the equation tan x = x in this interval.
(You may need to replot on smaller intervals to locate solutions.)

11

12

Introduction to Maple

Exercise 4. Graph
y = cos x

cos 3x
for x .
3

Then graph
y = cos x

cos 3x cos 5x
+
for x .
3
5

What happens if we continue this pattern that is, what does the graph of
y = cos x

cos((2k + 1)x)
cos 3x cos 5x
+
+ + (1)k
3
5
2k + 1

look like on [, ] as k tends to infinity?

Exercise 5. Graph the curve, defined by the polar equation r = 2(1 + cos ).
(This curve is called a cardioid.)

Exercise 6. The path of a point P on a circle of radius 2 that rolls along the
outside of a larger circle of radius 5 is given by

x(t) = 8 cos t 2 cos

7t
2

and

y(t) = 8 sin t 2 sin

7t
2

Draw the path of this particle for t in [0, 5]. Repeat for t in [0, 20]. (This curve
is called an epicycloid.)

Exercise 7. Some interesting curves, called Bowditch or Lissajous curves, are

given by coordinate functions of the form x(t) = cos at and y(t) = sin bt, for
constants a and b. Draw examples of these curves, experimenting with different
choices of a and b to get a feeling for what they look like in general. Start with
a = 3 and b = 5.

Exercise 8. Compare the graphs obtained from the two commands:

> plot( [2 + 3*sin(t), t, t = 0..2*Pi] );
> plot( [2 + 3*sin(t), t, t = 0..2*Pi], coords = polar );

1.4

1.4

Solving Equations

Solving Equations

Maple can also solve equations. Consider some examples:

> solve( x^2 + 3*x = 2.1 );
> solve( x^3 + x = 27 );
> solve( x^3 + x = 27.0 );
The second command gives an exact, but complicated, answer. Replacing 27 by
27.0 forces Maple to give decimal approximations instead, as do the commands
> fsolve( x^3 + x = 27 );
> fsolve( x^3 + x = 27, x, complex );
In general solve looks for exact answers using algebraic methods, whereas
fsolve uses numerical methods to find approximate solutions in floating-point
form. Compare
> solve( tan(x) - x = 2 );
> fsolve( tan(x) - x = 2 );
Notice how Maple responds if it cannot find the solution you asked for. Also,
notice that fsolve may not find all solutions. To understand why not, it is
helpful to look at a graph
> plot( { tan(x) - x, 2 }, x = 0..10, y = -10..10 );
This will give you some idea of how many solutions there are and what their
approximate location is. Then give fsolve a range of x-values in which to
search:
> fsolve( tan(x) - x = 2, x = 0..2 );
Often we need to use the solution of an equation in a later problem. To do
this assign a name to it. Here is one example.
> r := solve( x^2 + 3*x - 2.1 = 0 );
Note that the answer has the form r1, r2, where r1 is the first root and r2 is
the second. Such an object a bunch of items separated by commas, is called
an expression sequence. One picks out items of an expression sequence this way:
> r[1]; r[2];

13

14

Introduction to Maple

Here are some computations with items from an expression sequence:

> r[1] + r[2];
> r[1]*r[2];
> subs( x = r[1], 2*x + 3 );

# sum of the roots

# their product
# find 2(first soln) + 3

How did the sum and product relate to the coefficients of the equation?
We can also solve systems of equations:
> solve( { 2*x + 3*y = 1, 5*x + 7*y = 2 } );
> x; y;
A system of equations is given as a set a bunch of items enclosed in curly
brackets and separated by commas. Sets are often used when the order of the
objects is unimportant.
In reply to the solve command above, Maple tells us how to choose x and
y to solve the system, but it does not give x and y these particular values. To
force it to assign these values, we use the assign function:
> s := solve( { 2*x + 3*y = 1, 5*x + 7*y = 2 } );
> assign( s );
> x; y;
# Check that it worked.

Warning: You may have trouble later if you leave numerical values assigned
to the variables x, y, and r. Maple will not forget these assigned values, even
though you have gone on to a new problem where x means something different.
It is a good idea to return these variables to their unassigned state when you
finish your problem. Recall how to do this:
> x := x;

y := y;

r := r;

Recall also that restart clears all variables.

Finally, note that symbolic parameters are allowed in solve commands.
However, in that case we have to tell Maple which ones to solve for and which
ones to treat as unspecified constants:
> solve( a*x^2 + b*x + c, x );
> solve( a*x^3 + b*x^2 + c*x + d, x );
> solve( { a*x + b*y = h, c*x + d*y = k }, { x, y } );
The factor appearing in the denominator of the last computation is the determinant of the system of equations.
Exercise 1. Find as many roots of x5 + x = 32 as you can. How many do you
expect on theoretical grounds? You may want to see how fsolve works on this

1.4

Solving Equations

one. Here are some things to try:

>
>
>
>

solve( x^5 + x = 32 );
solve( x^5 + x = 32.0 );
fsolve( x^5 + x = 32, x );
fsolve( x^5 + x = 32, x, complex );

Exercise 2. Find all real solutions to x5 + x4 x3 x2 + 1 = 0. How many

are there?

Exercise 3. Using
> e1 := 7*x + 3*y + 8*z = 1;
solve({e1, e2, e3});

e2 := ... ;

e3 := ... ;

as a guide, solve the following systems:

(a)

7x + 3y + 8z = 1 (b)
2x 4y + 5z = 2
3x 7y + 9z = 1

7x + 3y + 8z = 1
2x 4y + 5z = 2

Exercise 4. Use solve to find the intersection points of the two ellipses defined
by
4x2 + 9y 2 = 1 and 25x2 + 4y 2 = 1.
Try using fsolve instead of solve. Finally, change the constant term in the
first equation to 1.0 and try solve again. Do the results agree? Draw a picture
to clarify the situation. You can do this by hand, or in Maple. The first ellipse
can be described parametrically by the expression ( 12 cos t, 13 sin t) where t runs
over the interval [0, 2]. The second ellipse has a similar parametric description.
By filling in the missing information below, you can plot both curves together:
> ellipse1 := [ cos(t)/2, sin(t)/3, t = 0..2*Pi ];
ellipse2 := [ ??? ];
plot({ ellipse1, ellipse2 }, scaling = constrained );

Exercise 5. Find all solutions to sin x + x = sec x for x in [3, 3]. Use plot to
get information about how many solutions exist and their approximate location.
Then use fsolve to locate the solutions to 9 decimal places accuracy.

15

16

Introduction to Maple

1.5

Functions

Although Maple has a large library of standard functions, we often need to

define new ones. For example, to define
p(x) = 18x4 + 69x3 40x2 124x 48
we say
> p := x -> 18*x^4 + 69*x^3 - 40*x^2 - 124*x - 48;
Think of the symbol -> as an arrow: it tells what to do with the input x,
namely, produce the output 18*x^4 + 69*x^3 - 40*x^2 - 124*x - 48. Once
the function p is defined, we can do the usual computations with it, e.g.,
>
>
>
>
>

p( -2 );
p( 1/2 );
p( 0.4 );
p( a + b );
simplify( " );

Warning: It is important to keep in mind that functions and expressions are

different kinds of mathematical objects. Mathematicians know this, and so does
Maple. Compare the results of the following:
>
>
>
>

p;
p(x);
p(y);
p(3);

#
#
#
#

function
expression
expression
expression

As further proof, try

>
>
>
>
>

factor( p );
factor( p(x) );
plot( p, x = -2..2 );
plot( p, -2..2);
plot( p(x), x = -2..2 );

Which of these worked? Does the factor command work on functions or expressions? What about plot?

1.5

Functions

Functions of several variables

Functions of several variables can be defined as easily as can functions of a
single variable:
> f := (x,y) -> exp(-x)*sin(y);
f(1, 2);
> g := (x,y) -> alpha*exp(-k*x)*sin(w*y);
g(1,2);
> alpha := 2; k := 1.3; g(1,2);
> w := 3.5; g(1,2);
> alpha := alpha; g(1,2);
Proc and unapply
There are two other ways to define functions. The first is with proc. For
example, to define f (x, y) = ex sin y, we say
> f := proc( x, y )
exp(-x)*sin(y);
end;
Note that there is no semicolon after the proc(x,y). This style is used most
often for functions that cannot be defined by a simple one-line expression, a
topic we will take up again in Section 8.
It is possible to define functions of zero arguments. For example,
> trial := proc()
rand(0..6)() + rand(0..6)();
end;
defines a function which simulates the operation throw two dice, then add up
the number of spots on the top faces. We test this:
> trial(); trial(); trial();
The second way to define a function is to use unapply. The code
> f := unapply( exp(-x)*sin(y), x, y );
converts the expression exp(-x)*sin(y) into a function of x and y. To verify
this, apply f to some some arguments:
> f(0,0); f(1.2, -0.4); f( u+v, arcsin(w) );
This last sentence reveals the secret of unapply: it is the reverse of the operation
of applying a function to arguments to obtain an expression. Riddle: what kind
of object is unapply( f(x), x )?

17

18

Introduction to Maple

A common use of unapply is to define a function within a function.

Exercise 1. The function
> rad := x -> evalf( x * Pi /180 );
converts degrees to radians. Thus rad(1) is the number of radians in one degree.
Modify this formula to define a function deg that converts radians to degrees.
Thus deg(1) is the number of degrees in one radian. Use these functions together
with Maples standard trig functions (which work in radians) to (a) find how
many degrees are in 1 radian, (b) to compute the sine of 1 , (c) to find an angle in
degrees whose cosine is 0.7, (d) to compute the height of a flagpole whose shadow
is 14 feet long when the suns rays make an angle of 11 with the vertical.

Exercise 2. Define a function dist of (u, v, x, y) that gives, in decimal form,

the distance from the point (u, v) to the point (x, y). Use your function to find
how far (11.34, 24.17) is from (9.61, 12.49).

Exercise 3. Define a function logb( b, x ) that computes the logarithm to

the base b of the number x, expressed in floating-point form. (Note that log(x)
in Maple means ln x.)

Exercise 4. Define f (x) = x + 1 and g(x) = 2x. Compute the composite

functions f g(x) = f (g(x)) and g f (x) = g(f (x)). Are they the same? Repeat
the computation in Maple. First define f and g and then
> (f@g)(x);
> (g@f)(x);
give the composite functions.

Exercise 5. Define f (x) = sin( x) using each of the three styles of function
definition. Use each definition to compute f (2) in decimal form.

1.6

Calculus

Let us now explore Maples tools for working with the notions of calculus.

1.6

Calculus

Derivatives
To compute the derivative of the expression x3 2x + 9, say
> diff( x^3 - 2*x + 9, x );
To compute the second derivative, say
> diff( x^3 - 2*x + 9, x, x );
or alternatively,
> diff( x^3 - 2*x + 9, x$2 );
This works because Maple translates the expression x$2 into the sequence x, x.
By analogy, x$3 would give the third derivative. Thus one can easily compute
derivatives of any order.
Now suppose that we are given a function g defined by
> g := x -> x^3 - 2*x + 9;
It seems natural to use
> diff( g, x );
to get the derivative g 0 . However, Maple expects an expression in x and so
interprets g as a constant, giving the wrong result. The command
> diff( g(x), x );
which uses the expression g(x), works correctly. The subtlety here is an important one: diff operates on expressions, not on functions g is a function while
g(x) is an expression.
To define the derivative of a function, use Maples D operator:
> dg := D(g);
The result is a function. You can work with it just as you worked with g. Thus
you can compute function values and make plots:
> dg( 1 );
> plot( { g(x), dg(x) }, x = -5..5,
title = g(x) = x^3 - 2x + 9 and its derivative );

19

20

Introduction to Maple

Partial derivatives
Maple can compute partial derivatives:
> q := sin( x*y );
> diff( q, x );
> diff( q, x, y );

# expression with two variables

# partial wrt x
# compute d/dy of dq/dx

As in the one variable case, there is an operator for computing derivatives of

functions (as opposed to expressions):
> k := (x,y) -> cos(x) + sin(y);
>
>
>
>
>

D[1](k);
D[2](k);
D[1,1](k);
D[1,2](k);
D[1](D[2](k));

#
#
#
#
#

partial with respect to x

partial with respect to y
second partial with respect to x
partial with respect to y, then x
same as above

Question: what is D[1,2](k) - D[2,1](k)?

Integrals
To compute integrals, use int. The indefinite integral (antiderivative)
Z
x3 dx
is given by int( x^3, x ). The following examples illustrate that Maple knows
integration by parts, substitution, and partial fractions:
>
>
>
>

int(
int(
int(
int(

1/x, x );
x*sin(x), x );
sin( 3*x + 1 ), x );
x / (x^2 - 5*x + 4), x );

Nonetheless, Maple cant do everything:

> int( sin( sqrt( 1 - x^3 )), x );
The last response contained an indefinite integral,
a signal that Maple does

not know how to find an antiderivative for sin( 1 x3 ) in terms of elementary

functions (the ones built from addition, subtraction, multiplication, division,
powers, roots, logarithms and exponentials, trig functions and their inverses).
In fact, it can be proved that no such antiderivative exists. Thus, even a smarter
Maple would not help.
To compute definite integrals like
Z

x3 dx
0

1.6

Calculus

we say int( x^3, x = 0..1 ). Note that the only difference is that we give an
interval of integration.
Let us return to the integral
Z

p
sin( 1 x3 ) dx

which Maple could not evaluate. We can still find an approximate numerical
answer using the definition of the integral: approximate the figure under the
graph by little rectangles or trapezoids and add up their areas. You could write
a loop to do this (see Section 8) or try:
> int( sin( sqrt(1 - x^3) ), x = 0..1 );
> evalf( " );
The second command forces Maple to apply a numerical method to evaluate the
integral. Of course, you could also put everything on one line:
> evalf( int( sin( sqrt(1 - x^3) ), x = 0..1 ) );
Numerical Integration
Another approach to numerical integration is to use the student package.
>
>
>
>

with( student );
j := sin( sqrt(1 - x^3) );
trapezoid( j, x = 0..1 );
evalf( " );

#
#
#
#

load package
define integrand
apply trapezoidal rule
put in decimal form

By default
the trapezoid command approximates the area under the graph of

sin( 1 x3 ) with four equal trapezoids. For greater accuracy use more trapezoids, i.e., a finer subdivision of the interval of integration:
> evalf( trapezoid( j, x = 0..1, 10 ) );
Better yet, use a more sophisticated numerical method like Simpsons rule:
> simpson( j, x = 0..1 );
> evalf( " );
Only an even number of subdivisions is allowed, as in simpson( j, x = 0..1,
10 ).
The student package is well worth exploring. Among other things it has
tools for displaying figures which explain the meaning of integration:
> leftbox( j, x = 0..1, 10 );

21

22

Introduction to Maple

The area of the figure displayed by leftbox is computed by leftsum:

> leftsum( j, x = 0..1, 10 );
> evalf( " );
One can also experiment with rightbox and middlebox and their companion
functions rightsum and middlesum.
Multiple integrals
Let R be the rectangular region defined by 0 x 1 and 0 y 1. To
compute the double integral
Z
(x2 + y 2 ) dxdy
R

in Maple, we compute the repeated integral

Z

(x2 + y 2 ) dx dy.
0

Here is one way to do this:

> int( x^2 + y^2, x = 0..1 );
int( ", y = 0..1 );
The first command integrates with respect to the x variable, producing an expression in y. The second command integrates the result with respect to y to
give a number.
Exercise 1. Compute
Z

(x2 + y 2 ) dy dx
1

The integrand is the same, but the order of integration is different. How do the
intermediate and final results compare with the previous computation?

Maple can compute double integrals over more complicated regions, such as
the triangle T defined by 0 x 1 and x y 1. We have

Z 1 Z 1
Z
2
2
2
2
(x + y ) dxdy =
(x + y ) dy dx
T

In Maple this translates as

> int( x^2 + y^2, y = x..1 );
int( ", x = 0..1 );
For more complicated integrands we can still use numerical methods, as
illustrated in the problems below.

1.6

Calculus

Exercise 2. Compute
Z

1
1

|x3 y 5 | dy dx

using Simpsons rule:

> j := sqrt( abs( x^3 - y^5 ) );
jy := simpson( j, y = -1..1 , 8):
jyx := simpson( jy, x = -1..1, 8 ):
evalf( jyx );

#
#
#
#

integrand
y integral
x integral
numerical result

Note the use of colons to suppress output. Repeat the computation with 16
subintervals. How accurate do you think the result is? Now reverse the order of
integration and compare answers.

Exercise 3. (Continuation) Compute the double integral

Z p
|x3 y 5 | dxdy
T

on the triangular region T defined by 0 x 1, x y 1 by computing the

repeated integral
Z
0

|x3 y 5 | dy dx.

Use Simpsons rule:

> jy := simpson( j, y = x..1, 8):
jyx := simpson( jy, x = 0..1, 8 ):
evalf( jyx );

# y integral
# x integral
# numerical result

Other Calculus Tools

Limits:
>
>
>
>

g := x
limit(
limit(
limit(

-> (x^3 - 2*x + 9)/(2*x^3 + x - 3);

g(x), x = infinity );
sin(x)/x, x = -infinity );
sin(x)/x, x = 0 );

Taylor expansions and sums:

>
>
>
>

taylor( exp(x), x = 0, 4 );
sum( i^2, i = 1..100 );
# Be sure i is clear
sum( x^n, n = 5..10 );
# needs n to be clear
sum( 1/j^5, j= 1..infinity ); evalf( " );

23

24

Introduction to Maple

Differential equations:
> deq := diff( y(x), x$2 ) + y(x) = 0;
> dsolve( deq, y(x) );
> ?dsolve
We can specify initial conditions and experiment with parameters.
>
>
>
>

k := k;
de := diff( y(x), x ) = k*y(x)*(1 - y(x));
sol := dsolve( {de, y(0) = 0.1}, y(x) );
k := 1; sol;

To graph the solution, we need only the right-hand side of sol, which we can
get by
> plot( rhs(sol), x = 0..10, title = k = 1 );
> k := 2; plot( rhs(sol), x = 0..10, title = k = 2 );
Exercise 4. Find the derivative of x5 + 7x2 + x + 1. Find the critical points
of f (x) = x5 + 7x2 + x + 1, and determine if they give maxima, minima, or
inflection points.

Exercise 5. Find an antiderivative for

both exact and floating point form.

1 + 7x2 . Compute

R1
0

1 + 7x2 in

Exercise 6. Find the Taylor series expansion of cos x centered at x = 0.

Compare the graph of the 4th order Taylor polynomial with the graph of cos x
itself on the interval [/2, /2].

Exercise 7. (a) Use the command sum to find a value for

1+

1
1
+ ... + 3 ...
3
2
n

that you believe accurate to three decimal places. How many terms do you need
to compute to get this accuracy?
(b) Repeat for
1

1
1
+ . . . + (1)n+1 . . . .
2
n

1.7

1.7

Vector and Matrix Operations

Vector and Matrix Operations

To work with vectors and matrices in Maple, first load the linear algebra
package:
> with(linalg);
Define and display a vector like this:
> v := vector( [1, -1] );
> v[1];
v[2];
v;
> print( v );
Note that Maple treats vectors as columns, even though it displays them as rows.
Now define another vector w and do some simple computations:
> w := vector( [1, 1] );
> add( v, w );
> dotprod( v, w );
What does this tell you about the geometry of v and w?
Next we define two matrices:
> A := matrix([ [2, 3], [1, 2] ]);
> B := matrix([ [1, 1], [0, 1] ]);
It is easy to make new matrices (or vectors) from old ones as in
> stack( A, B );
> augment( A, v, w );
Next, we do some computations with our matrices A and B and our vectors v
and w:
>
>
>
>

add(A, B);
scalarmul(A, 2);
multiply(A, B);
multiply(A, v + w);

Check to see if AB = BA:

> multiply(B, A);

#
#
#
#

or
or
or
or

use
use
use
use

evalm(A + B);
evalm(2*A);
evalm(A &* B);
evalm(A &* (v+w));

25

26

Introduction to Maple

Finally, we can change individual entries of a matrix with commands like:

> A[1,1] := 5; B[2,1] := -1;
> evalm(A), evalm(B);

# check

Maple can compute inverses and transposes:

>
>
>
>

inverse(A); # or use evalm( A^(-1) );

transpose(A);
inverse( transpose(B) );
transpose( inverse(B) );

It can also compute determinants:

> det(A);
> det( A + B ); det(A) + det(B);
> det( A &* B ); det( B &* A ); det(A) * det(B);
Symbolic matrices are as legitimate as numerical ones:
> A := matrix( [ [a, b], [c, d] ] );
> det(A);
Let
> C := matrix([ [3,2,2], [3,1,2], [1,1,1] ]);
We can put C into row-reduced form using
> gausselim(C);
or
> gaussjord(C);
There are also commands for doing elementary row and column operations on a
matrix: addrow, swaprow, mulrow, etc.
Exercise 1. Let v = (1, 2, 3) and w = (1, 4, 1). Find the sum and dotproduct
of v and w. Compute the length of v using the formula Length(v) = v v.
Exercise 2. The angle between two vectors v and w is given by the formula

vw
= arccos
|v||w|
where |v| stands for the length of v. Define a function Length that computes
the length of a vector v and a function Angle that computes the angle between
vectors v and w. Use it to find the angle between v = (1, 2, 3) and w = (1, 4, 1).
You may want to use evalf to get results in decimal form. Is your result in
radians or degrees? How would you modify your function definitions to get the
opposite?

1.8

Exercise 3. Let

A= 4
7

2
5
8

3
2

6
and B = 1
9
0

1
2
1

Programming in Maple

0
1.
2

Compute AB and BA. Are they the same? Compute A + B and B + A. Are
they the same? Compute 7A. Define C to be the sum of A and B. Then
compute C 2 and compare to A2 + 2AB + B 2 . Compute the determinants of
A and of B. Where possible, compute the inverses of A and B. Compute Av
where v = (1, 2, 3) as above. Is it possible to compute vA? Let x be an unknown
vector and solve Bx = v.

Exercise 4. Let A be the same matrix as in the previous exercise. Use augment
and gaussjord to solve Ax = v, where v = (1, 2, 3). Repeat for Ax = w, where
w = (1, 4, 1).

1.8

Programming in Maple

In this penultimate section we will take an extended look at the problem

of summing a series. It will serve as a vehicle for learning about the elements
of programming in Maple. We will study loops, conditionals, procedures, and
lists. Loops are for automating repetitive work. Conditionals are for choosing
between alternatives, i.e., for making decisions. Procedures are really the same as
functions, but the proc-style definition facilitates more complicated definitions.
Lists hold pieces of data and look like this: L := [ 2, 3, 5, 7 ]. The third
element of this list is 5, and we can form expressions like L[1] + L[2] to get
the sum of the first and second elements of the list.
Loops
Sometimes the quantity which we wish to compute cannot be gotten by
evaluating a simple formula. Consider, for example, the problem of computing
s(n) = 1 +

1
1 1
+ + ... + 2.
4 9
n

(8.1)

These are the partial sums of the infinite series

1+

1
1 1
+ + ... + 2 +
4 9
n

One way to compute them is to write out the defining expression for s(n) whenever we need to compute it:
>
>
>
>
>

1;
1 + 1/4;
1 + 1/4 + 1/9;
1 + 1/4 + 1/9 + 1/16;
etc.

27

28

Introduction to Maple

This works, but is not a good solution when n is large. We can cut down
somewhat on the amount of typing needed by judicious use of the ditto symbol,
which remembers the result of the last computation:
>
>
>
>

1;
" + 1/4;
" + 1/9;
" + 1/16;

The results will be more comprehensible if we work with decimal expansions

instead of fractions. We could accomplish this by writing
> 1.0;
> " + 1/4;
> etc.
The above method for computing s(n) is tedious and repetitive. We can
automate it with a loop. For example, to compute the first four terms of the
series (8.1) we say this:
> n := 4;
total := 0;
for i from 1 to n do
total := total + 1.0/i^2;
od;
First, we define n, the number of terms to add up. Next, we create a variable
total to hold the running total. Naturally, we set it to zero before we start.
Finally, we run the loop. It consists of two parts. The first, for i from 1 to
n, determines the number of times we do the loop. The variable i controls its
progress. The second part is the loop body, the phrase do total := total +
1/i^2; od . It consists of the key words do and od (that is, do spelled backwards)
enclosing a list of instructions. In our case we have only one instruction add
the quantity 1/i^2 to the current value of total and then store the result in
total.
Running our loop is the same as executing the sequence of commands below.
> n
i
i
i
i
i

:=
:=
:=
:=
:=
:=

4;
1;
2;
3;
4;
5;

total
total
total
total
total

:=
:=
:=
:=
:=

0;
total
total
total
total

+
+
+
+

1.0/i^2;
1.0/i^2;
1.0/i^2;
1.0/i^2;

Consequently, the counter i has the value 5 when the loop is complete. This
may cause trouble later, so you might want to clear the variable i with i :=
i when you have finished the problem.

1.8

Programming in Maple

Our loop as written prints out all the intermediate sums. For large n this is
not a good idea, so we rewrite our computation with colons to suppress display
of all but the final results:

> n := 10;
total := 0.0:
for i from 1 to n do
total := total + 1.0/i^2:
od:
total;

Warning: Only one Maple prompt (>) appears in the loop. This is the safest
way to type in a loop because it ensures that every time you start a loop, you
also end it. A command like

> for i from 1 to 5 do

>
i^2;
> od;

works, but can easily lead to disaster. For example, you might start typing a loop
and realize that you forgot to do something else first, say define a function that
you need inside the loop. Then you move back in your Maple file to take care
of this oversight. You have abandoned the loop, but Maple has not. Everything
you enter after a line like for ... do gets incorporated into the body of your
loop until you enter a line containing od to end the loop or until Maple crashes
and you lose all your unsaved work because Maple is irredeemably confused by
the nonsensical instructions it has been given. To avoid this problem type your
loops and procedures on a single Maple command line. To continue a command
on a new line without getting a new Maple prompt, you use Enter on a Unix
machine and Return on a Macintosh.

Conditionals
Next, consider the problem of finding partial sums of the series

1
1
1 1
+
+ ... 2 +
4 9 25
n

(8.2)

Depending on whether n is odd or even, we add or subtract the term 1/n2 . This

29

30

Introduction to Maple

can be accomplished using an if-then-else statement:

> n := 10;
total := 0.0:
for i from 1 to n do
if type( i, odd ) then
total := total + 1.0/i^2;
else
total := total - 1.0/i^2;
fi;
od:
total;
Note that the loop body now consists of more than one statement.
The general form of a conditional statement is
if <test> then
<something>
else
<something else>
fi;
The short form
if <test> then
<something>
fi;
is also legitimate.
Procedures
Suppose now that we need to compute partial sums many times in different
contexts. We would like to be able to say S(2), S(20), etc. instead of tinkering
with the loop above. To do so we define S as a function using the proc style
mentioned earlier.
> S := proc( n )
local i, total;
total := 0.0;
for i from 1 to n do
total := total + 1.0/i^2;
od;
total:
end;
Notice that the body of our definition the part between proc( n ) and end
is almost a word-for-word copy of what we wrote above to compute partial

1.8

Programming in Maple

sums. One important difference is that we declared the variables i and total
to be local. This means that they cannot be seen outside the function S, and,
moreover, that nothing that happens outside S can affect them. This kind of
protection is useful, since we cant keep in mind the names of all the internal
variables used by functions. The value of S(n) is the value of the last expression
in the definition, namely, total. One can also use the RETURN statement to
define the function value:
> S := proc( n )
local i, total;
total := 0.0;
for i from 1 to n do
total := total + 1.0/i^2;
od;
RETURN ( total );
end;

Warning: As with loops, it is much safer to type the entire definition with a
single prompt (>).
With a small bit of extra work, we can devise a procedure which computes
the nth partial sum
Sn = f (1) + f (2) + . . . + f (n)
where f is an arbitrary function. Here is the definition:
> S := proc( f, n )
local i, total;
total := 0.0;
for i from 1 to n do
total := total + evalf( f(i) );
od;
[ n, total ];
end;
The arguments of S are a function and a number. The function value S(f,n) is
a list consisting of two items: the number of terms in the sum, and the value of
the partial sum.
Warning: There is no punctuation after the phrase proc( f, n ). Also, with
procedures as with loops, it is much safer to type the entire definition on a single
Maple command line.

31

32

Introduction to Maple

Once we have typed in the definition of S, we can use it to recompute our

previous example 1 + 1/4 + 1/9 + 1/16. Define the function f
> f := x -> 1.0/x^2;
Then S( f, 4 ) gives the sum. Alternatively, one could say
> S( x -> 1.0/x^2, 4 );
To sum up a different series, we just change the arguments to S. For example,
> S( i -> 1/i, 10 );
sums the first ten terms of the harmonic series:
1+

1
1 1
+ + ... + + ...
2 3
n

Lists
Another way to think about the behavior of an infinite sum is to plot the
partial sums Sn versus n. This gives a picture of how the sum grows as n
increases. We can modify our procedure S to produce a list of points (n, Sn ) and
then use plot to graph them.
To define our list we use Maples sequence-building command. For instance,
to generate a list of squares we can type
> i := i;
# Make sure i is unassigned
> [ i^2 $ i = 1..10 ];
Whenever you use $ be sure that the indexing variable is unassigned or you will
get an error message. Alternatively, type
> [ i^2 $ i = 1..10 ];
to tell Maple to ignore any values previously assigned to the variable i when it
makes this list of squares. The quotation marks here are single quotes.
Let us call our modified procedure Spts. We define it like this:
> Spts := proc( f, n )
local i, j, total, pts;
total := 0.0;
for i from 1 to n do
total := total + evalf( f(i) );
pts[i] := [i, total]:
od;
[ pts[j] $ j = 1..n ];
end;

1.8

Programming in Maple

It produces a list of points of the form [k, Sk ] as we wanted. Now try

>
>
>

Spts( f, 10 );
Spts( f, 10 )[10];
plot( Spts(f, 10), style = line, labels = [n, Sn] );

Since we assigned n the value 10 earlier, this last command did not work. Try
again:
> n := n;
plot( Spts(f, 10), style = line, labels = [n, Sn] );
Now we have the desired plot. Replot using the first 100 terms of the series.
Does it look like the sum approaches a well-defined number as we take more and
more terms?
Exercise 1. Define a function f (x) = sin x/x. Investigate the limit of f (x) as
x 0 by computing the sequence f (1), f (1/2), . . . , f (1/10) with a loop.

Exercise 2. Define a function to compute the alternating series (8.2).

Exercise 3. (a) Use the procedures S and Spts to investigate the limit of the
partial sums
1
1
Sn = 1 + + . . . + 2
4
n
as n . Does the limit exist? That is, do the partial sums approach some
well-defined number or do they grow without bound?
(b) Repeat for the harmonic series
1+

1
1
+ ... + + ...
2
n

(c) Plot the partial sums for both series together:

> plot( {Spts(x ->1/x, 100), Spts(x->1/x^2, 100)});

Exercise 4. Study the alternating harmonic series

1

1
1 1 1
+ + ... + ...
2 3 4
n

Does it converge? If so, does it converge quickly or slowly? Compare with the
series
1
1
1 1
+ ... 2 + ...
1 +
4 9 16
n

33

34

Introduction to Maple

Exercise 5. Write a procedure S3 that sums the first n cubes. Notice that the
first five values, obtained by
> i := i;
> [S3(i) $ i = 1..5];
are perfect squares. Does this pattern continue? Modify your procedure to sum
the first n cubes, then take the square root of this sum, and give
i
h

n , sum of first n cubes , sum of first n cubes

as output. Compute the sequence [S3(i) $ i = 1..15]. What pattern do you
see? Does this pattern continue forever?

Exercise 6. Consider the sequence of odd counting numbers, grouped as shown

below:
1,

3, 5,

7, 9, 11,

13, 15, 17, 19,

21, 23, 25, 27, 29,

...

Notice that the first term is 1 , the sum of the next two terms is 8 or 23 , and the
sum of the next three terms is 27 or 33 . The sum of the next four terms is 64 or
43 . Does this pattern continue? Use Maple to check that this pattern continues
up to at least 153 . How could one be sure that it continues forever?

The next exercise contains a new kind of loop, the while-do loop. Its
general form is
while ( <test> )
do
<something>
<something else>
<etc>
od;
Exercise 7. Another variation on the procedures S and Spts is the following:
> G := proc( L )
local s, n;
s := 0.0;
n := 0;
while (s < L)
do
n := n+1;
s := s + 1.0/n;
od;
[s, n];
end;
Compute G(1.0), G(2.0), G(3.0), G(4.0), G(5.0), G(6.0), G(7.0). Try to
figure out what the procedure G is doing. What does it tell you about the
harmonic series 1 + 1/2 + + 1/n + ?

1.9

1.9

Troubleshooting

Troubleshooting

Although errors in Maple can be quite confusing at first, you will quickly
gain the experience needed to understand and fix them. Below is a list of common
errors to consider when troubleshooting.
1. Do statements end with a semicolon or colon?
2. Are parentheses, braces, brackets, etc., balanced? Code like { x, y } is
good but x, y } will cause trouble.
3. Are you trying to use something as a variable to which you have already
assigned a value? To clear a variable x, say unassign(x). You can
clear many variables at once, e.g., unassign(x, y). You will not have
to reload any packages. Remember, a variable that has a value assigned to
it no longer can function as a variable. To display the value of an ordinary
variable, type its name, followed by a semicolon, e.g.,
> x;
4. If things seem hopelessly messed up, issue the command
> restart;

5.

6.
7.
8.
9.
10.
11.
12.

You will have to reload any needed packages after this.

Are you using a function when an expression is called for, or vice versa?
Remember, ff := x^2 + y^2 defines an expression while f := (x,y) ->
x^2 + y^2 defines a function. It makes good sense to say f(1,3) but not
so much sense to say ff(1,2).
Are you using = when := is called for? Remember, the first tests for
equality while the second assigns a value to a variable.
Are you distinguishing between the three kinds of quotes? They are: ",
the double quote, , the single quote, and , the backquote.
In loops, do must be balanced by a subsequent od. In conditionals, if
must be balanced by a subsequent fi.
A procedure definition that begins with proc( ... ) must be be terminated by a subsequent end. There is no semicolon after proc(...).
If you use a function in a package, load the package first. Thus, to use
matrix, load linalg; to use display, load plots. You load linalg by
the command with( linalg ).
Do not confuse unassign with unapply.
Remember to use ? to inquire about the details of a Maple function, e.g.,
?plot for information about plot or just ? for general information.

35