84% found this document useful (19 votes)
9K views348 pagesAn Introduction To Linear Algebra by Krishnamurthy, Mainra & Arora PDF
Uploaded by
Yash GoyalCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
84% found this document useful (19 votes)
9K views348 pagesAn Introduction To Linear Algebra by Krishnamurthy, Mainra & Arora PDF
Uploaded by
Yash GoyalCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 348
An Introduction to
LINEAR ALGEBRA
V. KRISHNAMURTHY
Professor of Mathematics
: V. P. MAINRA
Assistant Professor of Mathematics
and
J. L. ARORA
Assistant Professor of Mathematics
All at the Birla Institute of Technology and Science,
Pilani
rye
AFFILIATED EAST-WEST PRESS PVT LTD
NEW DELHI - MADRAS Price Rs 22.50
‘No reproduction in any form of this book, in whole or in part
(except for brief quotations in critical articles or reviews),
may be made without written permission of the publishers.
Printed in India at Re&ha Printers,.New Delhi-110020
Published by Affiliate? East-West Press Private Limited
104 Nirmal Tower, 26 Barakhamba Road, New Delhi-110001 To Our Parents Contents
PREFACE
1 SETS AND FUNCTIONS
11 Sets J
1.2 Operations on Sets 5
1.3 Relations 8
1.4 Functions 10
1.5 Binary Operations 17
1.6 Algebraic Structures 20
17 Operations on Functions 24
VECTORS
2.1 Space Coordinates 33
2.2 Vectots—Addition and Scalar Multiplication
2.3. Dot Product of Vectors 48
2.4 Applications to Geometry 55
VECTOR SPACES
3.1 Vector Spaces 63
3.2 Subspaces 69
3.3 SpanofaSet 74
3.4 More about Subspaces 77
3.5 Linear Dependence, Independence 85
3.6 Dimension and Basis 95
4 LINEAR TRANSFORMATIONS
4.1 Definition and Examples 107
4.2 Range and Kernel of a Linear Map ‘114
43 Rank and Nullity 118
44 Inverse of a Linear Transformation 122
4.5 Consequences of Rank-Nullity Theorem 127
46 The Space L(U,V) 129
4.7 Composition of Linear Maps 133
48 Operator Equations 138
4.9 Applications to the Theory of Ordinary Linear .
Differential Equations 141
37
107 viii / CONTENTS
5 MATRICES
5.1 Matrix Associated with a Linear Map = 148
5.2 Linear Map Associated witha Matrix 154
5.3. Linear Operations in Mm n 158
5.4 Matrix Multiplication 163
5.5 Rank and Nullity of a Matrix 174
5.6 Transpose of a Matrix and Spec‘al Types of Matrices
5.7 Elementary Row Operations 185
5.8 Systems of Linear Equation, 195
5.9 Matrix Inversion = 200
6 DETERMINANTS
61 Definition 203
6.2 Fundamental Properties of Determinants 207
6.3 Proofs of Theorems 2/2
6.4 Cofactors 2/6
6.5 Determinant Minors and Rank of a Matrix 224
6.6 Product of Determinants 227
6.7 Applications to Linear Equations 230
6.8 Eigenvalues, Eigenvectors 233
6.9 Wronskians 238
6.10 Cross Product in V, 24]
7 MORE MATRIX THEORY
71 Similarity of Matrices 250
7.2 Inner Product Spaces 258
7.3 Orthogonal and Unitary Matrices 264
7.4 Application to Reduction of Quadrics 266
APPENDIX Ordinary Linear Differential Equations
ANSWERS TO PROBLEMS
INDEX
148
180
250
276
325 Preface
In Indian universities the emphasis at the undergraduate level has
been much more on calculus than on linear algebra and matrix theory.
Today, however, the need for linear algebra and matrix theory as an
essential part of undergraduate mathematics is recognised. The Bi-
national Conference in Mathematics Education and Research (June 1973)
recommended Elementary Linear Algebra as a compulsory course for all
students at the undergraduate level. It has since been generally agreed
that before a student begins to specialise in the discipline of his choice—
whether it be mathematics, science, engineering, social science, or
management—he must be exposed at least once to both calculus and
linear algebra; such an exposure will familiarise him with the concepts
and techniques of continuous mathematics (calculus), and the concepts,
methods, and logic of modern discrete mathematics (linear algebra).
This book is the outcome of a planned effort to teach linear algebra as
a second course in the mathematics curriculum introduced at the under-
graduate level several years ago at Birla Institute of Technology and
Science (BITS), Pilani The students who take this course have had a
semester of elementary calculus and analytical geometry. However, a
knowledge of the fundamental properties of continuous and differentiable
functions in terms of their addition, scalar multiplication, and multipli-
cation is sufficient for an understanding of this volume.
The treatment throughout is rigorous yet lucid. The fact that the
majority of students who would use this text may not ultimately become
mathematicians or physicists has not inhibited our development of the
subject. We strongly believe that present-day users of mathematics,
instead of being content with a hybrid of mathematical tools and
gymnastics which merely graze the subject, should delve deep by training
in concrete matter-of-fact arguments. The conceptual framework of linear
algebra and matrix theory provides the most efficient means for this
training; for, in one sense, matrices and linear equations form a concrete
foundation, and vector spaces and linear transformations give the flavour
of the abstract grandeur of modern mathematics. At the same time, as the
freshmen we are addressing may not have had any grounding in abstract X / PREFACE
mathematics, we have made a special effort to smoothen their first
encounter with methods of proof.
Theorems are proved in full (the end of proof is indicated by §), except
in rare cases where they are beyond the scope of the book. In these
instances the student is suitably instructed. Where certain consequences
of earlier results are stated as FACTS, the student will find he has been
sufficiently equipped in advance to prove them himself. The large number
of worked-out examples which have been woven into the text help the
Student to move back and forth from the concrete to the abstract. The
sets of problems—numerical, objective, and theoretical —interpolated at the
end of almost every article are a drill on the text. Answers to the
numerical problems appear at the end of the book; the objective questions
which are of the “‘true-false” type are intended to help the student in a
self-assessment of his conceptual understanding of the subject under
study.
Chapter 1 deals with sets and functions and gradually introduces the
language of modern mathematics. A teacher may adjust his pace in this
chapter to suit the standard of his class. Algebraic structures, such as
groups, rings, and fields, have been discussed only to the extent needed.
Chapter 2 provides the concrete geometric structure of 2- and 3-dimen-
sional vector spaces. It leads the student to the problems of geometry
through vectors and prepares the ground for Chapter 3 which gets into
the essence of the subject. Here the theory of vector spaces, and the con-
cepts of linear dependence and linear independence, dimension and basis
are treated elaborately. Though infinite-dimensional vector spaces are
also considered, the emphasis throughout is on finite-dimensional vector
spaces.
Chapter 4 aims to familiarise the student with the fundamental pro-
perties of linear transformations. The rank-nullity theorem and its
consequences are presented in deta'l. The theory developed so far is
applied to operator equations and, in particular, to differential equations.
This application discloses that the solution space of the m-th order normal
homogeneous linear differential equation is an n-dimensional subspace of
the space of n-times continuously differentiable functions. The further
application to the theory of ordinary linear differential equations is detail-
ed in the Appendix. However, we have not attempted to make the
treatment of differential equations self-contained.
The elaborate build-up on vector spaccs and linear transformations
begins to pay dividends in Chapter 5 which starts by establishing the tink
between linear transformations and matrices. In our experience, the
welter of mathematical detail on matrices in this chapter is easily assimi-
lated by the student because of the knowledge of linear transformations
hehas gained in Chapter 4. He is thrilled to see that the elementary PREFACE | xi
(apparently trivial) row operations on matrices finally result in the solution
of linear equations in all their ramifications. Naturally, the chapter ends
with matrix inversion.
Now the student is ready for determinants, presented in Chapter 6.
When he comes to determinant minors and the rank of a matrix he realises
the importance of the emphasis in Chapter 3 on the concept of linear
dependence and linear independence. The theorem giving the connection
between the rank of a matrix (already defined in Chapter 5 by means of
independence concepts) and the order of its nonzero minors is the crux of
the content here. The ease with which it is proved justifies the efforts
taken in the development of the subject in earlier chapters. Applications
to linear equations, and a brief account of eigenvalues and eigenvectors of
matrices, Wronskians, and the cross-product in V, give an idea of what
determinants can do.
Chapter 7 gives a glimpse of the theory of orthogonal and unitary
matrices, similarity transformations and their application to the geometry
of quadrics. When the student reaches this chapter, he easily recognises
the connection batween linear algebra and gcometry.
The student should guard against conceptual errors of three types :
‘finite dimension’ versus ‘infinite dimension’; ‘real scalar’ versus ‘complex
scalar’; and ‘non-empty set’ versus ‘empty set’. When in doubt regarding
hypotheses he should invoke what may be culled an ‘emergency axiom’ :
‘The suitable alternative in each relevant pair(s) is included as an additional
hypothesis.
The text can be adapted to suit different curricula : as a one-year
course running three hours a week; as a one-semester course running five
or six hours a week (as is the practice at BITS); or, by a judicious selec-
tion of topics, as a onesemester course running three hours a week. It can
be used during any year at the undergraduate level or as part of a first
course in linear algebra at the postgraduate level, if no matrix theory has
been done till then. However, as the topics have been arranged sequen-
tially, any student who wishes to change the order of topics will need
guidance. :
We wish to thank Dr. C. R. Mitra (Director, BITS) and others at BITS
who encouraged our efforts in writing this book under the Course
Development Scheme and provided all the necessary assistance in dupli-
cating and class-testing its earlier versions over a period of thrée ycars. To
the BITS students—about 1200—of these three years, we are alsé. indebted
for their lively response to our experiments in pedagogy.
We are grateful to Messrs. G. R. Verma and B. L..Soni who patiently
typed the several drafts; Mr. A. P. Rastogi who drew the illustrations; and
the National Book Trust, India, for subsidising the publication of the xii / PREFACE
book. To Mrs. J. Fernandes of East-West Press, our appreciation and
thanks are due for a thorough editing of the manuseript.
Our last words of affectionate gratitude are reserved for our wives and
family members who continually cheered our years of effort at the project.
We shall welcome all suggestions for improvement of the book.
V. Krishnamurthy
Pilani V. P. Mainra
September 1938 J. L, Arora Chapter 1
Sets and Functions
We shall begin with a discussion on ‘sets’. Immediately, we get into
the first requirement of mathematics, viz. proper definitions for
technical terms. A major concern of mathematics is precision not only
in the calculations of quantitative information, but also in the communi-
cation of thought. This is why importance is given to definitions in
mathematics. In this book we shall come across a large number of
definitions. Each definition will introduce a new technical term or a new
concept in terms\of preceding definitions. The very comprebension
of a definition may often depend on the logical development of the subject
up to that point.
However, the first technical term, namely, ‘set’ will be introduced
without a precise definition. The reason is obvious: The moment we
attempt to define ‘set’, we get into words such as ‘collection’ or ‘aggregate’,
which themselves need to be defined. We have to draw the line some-
where !
it SETS
The meaning of ‘set’ is intuitively understood as a well-determined
collection of objects, called its ‘members’ or ‘elements’. The objects
(members or elements) are said to ‘belong to’ the set or to be ‘in’ the set.
Here all the words in quotation marks are taken to be undefined terms.
To illustrate the meaning of ‘set’, let us consider some examples.
Example 1.1 The collection of the three boys : Arun, Moban, and Ram.
Example 1.2 The collection of the three symbols: A, 0 and Q.
Example 1.3 The collection N of all natural numbers.
Example 1.4 The collection Z of all integers.
Example 1,5 The collection Q of all rational numbers.
Example 1.6 The collection R of all real numbers.
Example 1.7 The collection of all the past presidents of India.
Example 1.8 The collection of all the first year students of Birla Institute
of Technology and Science (BITS). 2] Sets AND FUNCTIONS
Example 1.9 The aggregate of the living men in the world whose height
exceeds 2 metres.
Example 1.10 The aggregate of the roots of the equation x*”7 — 1 = 0.
Example 1.11 The aggregate of the integers that leave a remainder 2
when divided by 5.
Example 1.12 The group of cricketers who were out for 99 runs in a
test match.
Example 1,13 The collection of all positive primes.
Example 1.14 The collection of derivatives of all orders of the function
e
Example 1.15 The collection C of all complex numbers.
All these are examples of sets. We can construct several such exam-
ples. Let us now consider two collections which are not sets :
- (i) The collection of some natural numbers,
(it) The collection of the politicians of India.
In (+) it is not clear which numbers are included in the collection.
Whether the number 2 is in the collection or not cannot be answered
without first explaining the word ‘some’. Again, in (1i) the question
whether a specific person is a politician or not would get different responses
from different persons. Thus, collections (i) and (ii) are not ‘well deter-
mined’.
In contrast to these two examples it is worthwhile to analyse Examples
1.12 and 1.13. In Example 1.12 we note that every cricketer is either in
the group or he is not in the group. We do not have to check any records
to say this. In Example 1.13, again, either a number is a prime or it is not
aprime. Here it matters little whether it is known that a particular number
is a prime or not. In other words, it is immaterial whether we can
answer the question: Is this particular object in the given collection or
not? What matters for specifying a set is to know unambiguously that
only one of the two answers is possible : The object’in question belongs
to the given collection or it does not belong to it.
Thus, we can elaborately describe a set as a collection of objects
which is well determined in the sense that, for every object, there should
be only two possibilities available unambiguously : either it belongs or it
does not belong to the collection.
If A is a set and x is an element of A, then we write x € A (€ is read
as "belongs to’ or ‘is in’ or ‘is an element of” or ‘is a member of’). The
negation of this is denoted by x & A (& is read as ‘does not belong to’ or
‘is not in’ or ‘is not an element of” or ‘is not a member of’).
Jastead of such a detailed description of sets, two kinds of symbolic
descriptions are very often used. One is by listing, if possible, all the 11 sers / 3
elements of the set within braces, e.g. sets of Exumples 1.1 and 1.2 are
respectively written as :
{Arun, Mohan, Ram}
and {A, 0, 9}.
The other is by using a dummy element, say x, and writing the character-
istic properties of x which precisely make it an element of the set. Thus,
the set of elements characterised by the properties, say P, Q, ..., is written
as
{a | x satisfies P, Q, ...}
or {x rx satisfies P,Q, ...}.
(‘| ° and *:* are reed as ‘such that’.) Thus, the sets in Examples 1.11 and
1.12 may respectively be written as
{x | x is an integer and x = 5k + 2 for some integer k}
and a. o is a cricketer who was out for just 99 runs
‘ina test match
Note that in the first method the order in which the elements are listed
is immaterial. Thus, {Arun, Mohan, Ram} and {Mohan, Arun, Ram} are
the same sets, We shall make this precise by defining the equality of
sets.
1.1.1 Definition Two sets A and B are said to be equal if every member of
A is a member of B, and every member of Bis a member of A. In
such a case we write A = B. .
For example, {0, 1, 2, 3} = (2,1, 0,3}. Also Z = {x|x is an in-
teger} = {..., —3, —2, -1,0, 1, 2, 3, ...}.
1.1.2 Convention All our definitions are ‘if and only if’ (if*) statements.
For example, if a definition reads ‘A triangle is said to be equilateral
if all its sides are equal’, we actually mean that ‘A triangle is said to
be equilateral if all its sides are equal’. :
Torview of. this convention, Definition 1.1.1 means that two sets A and
B are said to be equal iff every member of A is a member of B, and every
member of B is a member of A. .
SUBSETS
Let A and B be two sets such that every member of 4 is also a member
of B, Then A issaid to be a subset of B. In symbols we write 4 C B
(‘C’ is tead as ‘is a subset of? or ‘is contained in’), Whenever A is. a
subset of B, we also say that Bis a superset of A. In symbols we write
BD A(‘D’ is read as ‘is a superset of’ or ‘contains’). * .
Obviously, every set is a subset (superset) of itself. 4 | SETS AND FUNCTIONS
Exercise Prove that A = BiffA C Band BC A.
This exercise is immediate and the reader can prove it himself. In
Practice, whenever we want to prove that two sets A and B are equal, we
prove both the inclusions, ie. AC Band BC A.
EMPTY SET
Consider the set of women presidents of India elected before
31st December 1974. There was no such woman president. So this set
has no members in it. Such a set, i.e. a set containing no elements in it,
is called an empty set or a null set. It is denoted by 4.
It may be noted here that there is only one empty set. As a clarifica-
tion of this, note that the set {x : x is a real number satisfying x* + 1 = 0}
is also empty and we can write,
The set of all women presidents of India elected before
31st December 1974 =>
= {x: x is a real number satisfying x* + 1 = 0}.
The set {x : x is a real number satisfying x* + x — 0} consists of only
one member, namely, zero. So this set can be written as {0}. Note that this
is not the empty set.
Nonempty sets that consist of only a single member are called singletons.
The set in Example 1.14 is a singleton as itconsists of only one element e*.
It would be interesting (at least for cricket fans) to find out whether the
set in Example 1.12 is empty or a singleton or has more members than
one.
Finally, we note that @ C A for all sets 4. Thus, given a set A, it has
two extreme subsets. One is A itself and the other is ¢. Any subset
of A other than A and ¢ is called a proper subset of A.
Problem Set 1.1
1. Let 4, B, and C be three sets such that A C Band BC C. Then prove
thataA CC.
2. Let Sy, Sy, ..., S, be n sets such that 9, C Sy C ... C S, and S, C Sy.
Then prove that S; = S, = «+. = Sy.
3. Prove the exercise given on page 3.
4, Determine the set of real numbers x satisfying the following :
() #<1 &) x + <0 @ Zt <0
@) 2-430 (©) G4+2I@—N)S0M «+1
