CHAPTER

Linear Equations
and Matrices

m Systems of Linear Equations

. The equation
(I )

which expresses the real or complex quantity b in tcrms of the unknowns X I , X2,
.. • , X" and the real or complex constants (II. (12 • ... • (In, is called a linear equa-
tion. In many applications we are given b and must find numbers Xl. Xl . ... . X"
satisfying (I).
A solution to Iinea.r Equation (I) is a sequence of /I numbers SI.,f2.. .,f".
which has the propeny that ( I) is satisfied when XI = ,fl, -'"2 = S2, . .. ,X" = s" are
substituted in (1). Thus XI = 2, X2 = 3. and XJ = - 4 is a solution to the linear
equation

because
6(2) - 3(3) + 4( - 4) ~ - 13.
More generall y, a system of III linear equations in /I unknowns, Xl, -'"2 .. . .. X".
or a linear system, is a set of III linear equations each in n unknowns. A linear

I
2 Chapler 1 Linear Equations and Matrices

system can conveniently be wri tten as

(2)

Th us the ith equation is

I.n (2) the (Ii) are known constants. Given values of hi. h 2 • ...• h",. we want to find
values of XI. X2 . .. .. x " that will satisfy each equation in (2).
A solution to linear system (2) is a sequence of 1/ numbers SI. S2. . . SII'
which has the property that each equation in (2) is satisfied when XI = .1"1. X2 = .\·2 ,
.... x" = .f" are substituted .
If the linear system (2) has no solution, it is ~aid to be inconsistent: if it has
a solution. it is called consistent. If b l = h2 = = b", = 0, then (2) is
called a homogeneous system. Note that XI = x? = = x" = 0 is always
a solution to a homogeneous system; it is called the trivial solution. A solution
to a homogeneous system in which not all of XI. X2 • ... • XII are zero is called a
nontrivial solution.
Consider another system of r linear equations in 1/ unknowns:

CIIXI + CI2X2 + ... + Cl n T" = til

C21XI + CnX2 + ... + Cil,T" = til

(3)

We say that (2) and (3) are equivalent if they both have exactly the same solutions.
The linear system
EXAMPLE 1
XI - 3X2 = - 7
(4)
2xI + X2 = 7

has only the solution XI = 2 and X2 = 3. The linear system

8x, - 3X2 = 7
3x, - = 0
2tl (5)
IOXI - 21:2 = 14

also has only the solution XI = 2 and X2 = 3. Thus (4) and (5) are equivalent. •
To find a solutio n to a linear system, we shall use a technique called the
method of elimination:
 1.1 Systems of linear Equations 3

Consider the linear system

EXAMPLE 3
x -3)"=-7
(7)
2x - 6y = 7.

Agai n. we decide to eliminate x. We add (-2) times the fi rst equation to the
second one, obtai ning
0 = 21.
which makes no sense. This means that (7) has no solutio n: it is inconsistent We
could have come to the same conclusion from observing that in (7) the left side of
the second equation is twice the len side of the fi rst equatio n, but the right side of
the second equation is not twice the ri ght side of the fi rst equation . •
4 Chapler 1 Linear Equations and Matrices

Consider the linear system

EXAMPLE 4
x + 2)' + 3z = 6
2x - 3)' + 2z = 14 (8)
3x + )' - z = - 2.

To eliminate x, we add (- 2) times the first equation to the second one and (- 3)
times the fi rst equation to the third one, obtaini ng

- 7)' - 4 z= 2
(9)
- 5)' - IOz = - 20.

Th is is a system of two equations in the unknow ns.v and z. We multiply the second
equation of (9) by (- tl. yieldi ng

- 7y - 4z= 2
)' + 2z = 4,
which we write. by interchanging equations. as

y + 2z= 4
( 10)
- 7y - 4z= 2.

We now eliminate y in ( 10) by adding 7 times the first equation to the second one,
to obtai n

10z = 30.

z= 3. ( II )

Substituting this value of z into the first equation of (10), we find that y = - 2.
Then substituting these values of y and z into the first equation of (8), we find that
x = I. We observe further that our elimi nation procedure has actually pnxluced
the linear system

x + 2y + 3:; = 6
)' 1 2;: = 4 (12)
:; = 3,
obtained by using the first equations of (8) and (10) as well as (II). The importance
o f this procedure is thaI, although the linear systems (8) and (12) are equi valent.
(12) has the advantage that it is easier to solve. •

Consider the linear system

EXAMPLE 5
x + 2)' - 3z = - 4
(13)
2x + y - 3z = 4.
 1 .1 Systems of linear Equations 5

Eliminating x, we add (-2) times the first equation to the second equation to get

-3)' + 3z = [ 2. ( 14)

We must now solve (14). A solution is

)' =z - 4.
where z can be any real number. The n from the fi rst equation of ( [ 3),
x = -4 - 2y + 3;:
= - 4 - 2(z - 4) + 3z
= z + 4.
Thus a solution to the linear system (13) is

x = z +4
Y= z- 4
z = any real number.
This means that the linear system (13) has infinitely many solutions. Evcry time
we assign a value to z we obtain another solution to ( 13). Thus, if z = I, then

x = 5. y = - 3. and

is a solution, while if z = - 2. then

.r = 2, y = - 6. and z = -2

is another solutio n.
•
These examples suggest that a li near system may have a unique solution, no
solution, or infi ni tely many solutions.
Consider next a li near system of two eq uations in the unknowns x and y:

(JIX + (J2)' = el
(15)
blx + b 2)' = C2 ·

The graph o f each o f these equations is a straight line. which we denote by il and
£2, respectively. If x = .\'1,)' = .f! is a solution to the linear system (15). then the
point (.1'1. S2) lies on both lines i l and i 2 . Conversely, if the point (.fl. Sl) lies on
both li nes i l and £2, then x = .1"[, )' = .\. ! is a solution to the linear system (15).
Thus we are led geometrically to the same three possibilities mentioned previously.
See Figure I. [.
Next. consider a linear system o f three equations in the unknowns x. y, and z:

([IX + b l )' + elZ = d l

(J2X + b 2 J' + elZ = d 2 ( 16)
(l3X + h3)' + e3Z = d 3 .

The graph of each o f these equations is a plane. denoted by PI. P2, and P3, re-
specti vely. As in the case of a linear system of two equations in two unknowns.
6 Chapler 1 Linear Equation s and Matrices

(a) A unique solution. (b) No solution. (e ) Infinitely many solutions.

)' )' )'

I,
I,
- 7"+---'__- < ---\1--\--" ---+7"'---- "
I, I,
FIGURE 1. 1

the linear system in (16) can have infini tely many solutions, a unique solution, or
no solution. These situations are illustrated in Figure 1.2. For a more concrete
ill ustration of some of the possible cases, consider that two intersecting walls and
the ceiling (planes) of a room intersect in a unique point, a corner of the room,
so the li near system has a unique solution. Next, think of the planes as pages of
a book. Three pages of a book (held open) intersect in a straight line, the spine.
Thus, the linear system has infi ni tely many solutions. On the other hand, when the
book is closed, three pages of a book appear to be parallel and do not intersect, so
the linear system has no solution.

(a) A unique solution. (b) Infinitely many solutions. (c) No solution.

p,/ P,

V
./
p;/
FIGURE 1.2

If we exami ne the melhod of elimination more closely, we fin d that it involves

three manipulations that can be perfonned on a linear system to convert it into
an equivalent system. These manipu lations are as follows:
I. Interchange the ith and Jth equations.
2. Multiply an eq uation by a nonzero constant.
3. Replace the ith equation by c times the jth equation plus the ith equatio n,
i i= j. That is, replace

by
8 Chapler 1 Linear Equation s and Matrices

Key Terms
Linear equation Consistent system Unique solution
Solulion of a linear equation Homogeneous system No solution
Li near system Trivial solution Infinitely many solutions
Unknowns Nontri vial sol ution Manipulations on linear systems
Inconsistent system Equivalent systems Method of elimination

M.M Exercises

I II EJerdse.\· Ilhro l/gh 14. .\"O/re each gil'ell lillear ~)'~'Ielll by 3. 3x + 2y + : = 2 4 . .{ +y =5

fhe It1nl/Od of dimillmivil. 4x + 2y + 2: =8 3x+3y=IO
x - y+ : =4
l. x + 2)' = 8 2. 2x-3y+4;= - 12 5. 2{+4y+6Z = - 12 6. x+ y-2: =S
h - 4y =4 x - 2y + : = -5 2{ - 3y-4~ = 15 2x + 3y + 4: = 2
3x+ y+2z = 3x+4y +5: = -8
 1 .1 Systems of linear Equations 9

7. .1 + 4y - :: = 12 8. 3x + 4)' - z =8 (a ) Verify that Xl = I. Yl = - 1. Zl = - I is a solution.

h + 8y - 2;:: = 4 6_1 + 8)' - 2;:: = 3 (b) Verify that X2 = - 2.)'2 = 2. Z2 = 2 is a sol ution.
9. x + ), +3z =12 10. x+ )' =1
2x + 2)' + 6;:: = 6 2, - ), =5 (e) Is x = X l + X2 = - I. Y = Yl + Y2 = I. and
Z = Zl + Z2 = 1 a solution to the linear system?
3x + 4)' = 2
II . h+3y=13 12. x - 5)' = 6 (d ) Is 3x. 3y. 3z . where x. y. and; are as in part (c). a
x - 2)' = 3 1x + 2)' = I solution to the linear system?
5:.: + 2)' = 27 5.{ +
2)' = I 20. Without using the method of elimination . so lve the linear
13. x + 3.1' = - 4 14. 21 + 3)' - z = 6 system
2x+5y=-8 2, - ),+2z = - 8 2x + )" - 2z = - S
x+3y=-S 3x - y+ z = - 7 3y + z = 7
15. Given the linear system 4.
2x - ) =5 21. Without using the method of el imination. solve the linear
4x -2) = /. system
4x 8
(a ) Delennine a particular vahle of I so thallhe system
is consistent. -2\" + 3)' = - 1
(b) Detennine a particular value of f so thallhe system 3x+Sy - 2z = II.
is mconslstenl. 22. Is there a value ofr so that X = I. Y = 2. Z = r is a
(c) How many different values of I can be selected in sol ution to the followin g linear system? If there is. find
pari (b)? it.
16. Given the linear system 2x + 3)' - Z = I I
x - y+2:=-7
3x+4)=s 4x+ y-2z = 12.
6x+8) =I.
23. Is there a val ue ofr so that x = r.y = 2. : = lis a
(a) Determine particular values for l' and I so Ihal the
sol ution to the follol'.-ing linear system? If there is. find
system is consistent.
it.
(II) Dtlennine p.uticulal values fOJ 1 and I so thai the 3x - 2z= 4
system is inconsistent.
x- 4)' + z =-S
(c) What relationship between the values of s and I will
-2x + 3y + 2z = 9.
guarantee Ihat Ihe system is consistent?
17. Given the linear system

x+ 2y=10
3x + (6+1»), = 30.
(a ) Determine a particular value of I so that the system
has infinitely man y solutions.
(b ) Determine a particular value of I so that the system
has a unique solution.
(e) How m:my different values of I can be selected in
part (b)?
18. Is every homogeneous linear system always consistent?
Explain.
19. Given the linear system

2.1"+3), - z =O
x - 4)' +5z =0.
 1.2 Matrices 11

m Matrices

A n 11/ X 11 matrix A is a rectangular array of mil real or complex numbers arranged

DEFINITION 1.1
in //I hori zontal rows and II vertical columns:
a!! al2
a2! a22

(I)
.....- ith row

•L- jth column

The ith row o f A is

ai ll ] (I :;: i '=:: /I/ );

thejth column of A is

aa,'i] (l .=::J.=:: n).

[ ,L
We shall say that A is III by II (written as III x II ). If 11/ = II , we say that A is a
square matrix of order II , and thai the numbers II!!. a22 . .... all" form the main
diagonal of A. We rc fer to the number a ij , which is inlhe ith row and Jth column
o f A, as the i,j th element of A, or the (i.j) entry of A. and we often write (I) as
12 Chapter 1 Li near Equations and Matrices

Lei
EXAMPLE 1

A= [ [
- \
2
0 n B= [ I+ i
2 - 3i
4;
- 3 .
1 C ~ Hl
D~ [i 0
- \ n E ~ [3]. F ~ [- \ 0 2] .

T hen A is a2 x 3 matriK witha l2 = 2,{/ u = 3,(122 = 0, and un = [; B is a2 x 2

matrix withb ll = I +i,bl2 = 4i. b2 1 = 2 - 3i.and b n = - 3; e is a3 x I matrix
with C I I = L C2 1 = - I, and C3 1 = 2: D is a 3 x 3 matrix: E is a I x [ matrix:
and F is a [ x 3 matrix . I.n D, the elements dll = l. d 22 = 0, and d 3J = 2 form
the mai n diagonal. •

An /I x I matrix is also called an lI·vector and i:; denoted by lowercase boldface

letters. When II is understood, we refer to II-vectors merely as vectors.

EXAMPLE 2 •
T he II-vector all of whose entries are zero is denoted by O.
Observe that if A is an II x /I matrix, then the rows of A are I x II matrices and
the columns of A are II x I matrices. The set o f all /I-vectors with real entries is
denoted by R". Simi larly, the set of all II-vectors with complex entries is denoted
by en.
14 Chapter 1 Linear Equations and Matrices

Two III x /I matrices A = [ai) ] and B = [hi) ] are equ al if they agree entry by
DEFINITION 1.2
entry. that is, ifaij = bij fori = 1.2, . .. . 111 and j = 1. 2 . .. . . 11.

The matrices
EXAMPLE 7

~ [~ ~]
2
B x
)' -4

are equal ifandonly ifw = - I, x = - 3,y = O, and z = 5.

•
 1.2 Matrices 15

• Matrix Operations

Matrix Addition
If A = [aij ] and B = [bij ] arc both 11/ x /I maITiec;, then thc s um A + B is an m xn
DEFINITION 1.3
matrixC = [Ci j ] dcfined by Ci j = aij+ hij .i = L2 .. .. . m;j = 1. 2 . . 11.
Thus, to obtain Ihc sum of A and B, we merely add corresponding entries.

Lei
EXAMPLE 8
A = [~
- 2
-I !] ond B= [~
2
3 -~]
Then

A+B =
[I +0 - 2+ 2 3+
2+ 1 - I + 3 4 + (- 4)
1] [I = 3
0
2 ~]

If x is an II-vector. then it is easy to show that x + 0 = x, where 0 i, the

II-vector all o f whose entries are zcro. (See Exercise 16.)
It should be noted that the sum of the matrices A and B is defined only when
A and B have the same number of rows and the same number of columns, that is,
only when A and B are o f the same size.
16 Chapter 1 Linear Equations and Matrices

We now make the convention that when A + B is written, both A and B are of
the same size.

Scalar Multiplication
If A = [aij ] is an m x /I ma trix and r is a real number, then the scalar multi ple of
DEFINITION 1.4
A by r, rA. is them x /I mat ri x C = [cij J. whe recij = rai j , i = 1. 2 . .. .. II! and
j = 1. 2 . .. .. II ; that is, the mat ri x C is obtained by multiplying each entry of A
by r.

We have
EXAMPLE 10
_, [4
- 7
- 2
- 3
- 3]
2
~ [ (-2)(4)
( - 2)(7)
(- 2)( - 2)
(- 2)( - 3)
(-2)(- 3)]
(- 2)(2)

=
[ -8
- [4
4
6 -:1
Thus far, addi ti on of matrices has been dcJined for only two matrices. Our
•
work with matrices will call for addi ng more than two mat rices. Theorem 1.1
in Section 1.4 shows that addit ion of matrices satis fies the associative propert y:
A + ( 8 + C) ~ (A + B)+ C.
If A and B are 11/ x II matrices. we wri te A + (- 1) 8 as A - 8 and call thi s
the difl'erence between A and B .

Lei
EXAMPLE 11
A= [~
3
2
and 8~ [2 3
- I
5
Then

A_ B= [2 - 2 3+ 1 4 -8]
3 .
4- 3 2- 5 -3
•
 1.2 Matrices 17

If A I. A 2 • ...• Al are 11/ x 1/ matrices and Ct . C2 • .. . • Ct arc reat numbers. then

an expression of the foml
(2)

is calted a linear comhinalion of A t. A 2 •.. .• A.., and Cl . C2 • .. .•q are called

coerticients.

T he following are linear combi nalions of matrices:

EXAMPLE 13

3
0
2
[ I
- 3
3 5] 2I[ 5
4 - 6
2
2
- 2 - 3 - I -2

2[3 - 2] - 3[5 0] + 4[ - 2 5].

18 Chapter 1 Linear Equations and Matrices

Using scalar multiplication and matrix addition, we can compute each o f these
linear combinations. Verify that the results of such computations are, respectively,

-10
[-; 8 ""2¥]. [-17 16].
-5 - ""2
ond
[-046]
0.4 .
3.08
•

DEFINITION 1.5
If A = J,
[aij ] is an 11/ x II matrix, then the t r anspose of A. AT = [a~ is the
n x matrix defined by {I~ = {I jl ' Thus the transpose of A is obtained from A by
11/

interchangi ng the rows and columns o f A.

Lei
EXAMPLE 15

[~ [ -~ ~].
-2
A = 5
2 - 3
 1.2 Matrices 19
T hen

H~l U n
3
AT = BT = - I
- 2 2

C
T
= [~ -3
2
-n DT =
Hl and £T = [2 - I 3]'

•
Key Terms
Matrix EqU3J mmrices Difference of matrices
Rows /l-vector (or vector)
Columns Rn , c n
Size of a matrix 0, zero vector Linear combination
Square matrix Coefficients
Main diagonal Matrix addition Transpose
Element or entry of a matrix Scalar multiple

NU Exercises
l. LeI

A = [~
and

n
4. If
c~
["c -+b
d
<+d]
(/ - b =
[4 [0

(a) Whatisa I2.an.(I2l'! find a, b, (". and d.

(b) Whatish ll .b31 '! 5. I f
(e ) Whatis c 13.C31 .(:31 ? [a + 2b
2c + d
2a - bH4
c - 2d 4
-2]
-3 .
find a, b, c . and d.

III £xercise~' 6 rhrvugh 9, lei

A = [~
2
:1 B~ [i n
~ [~ -2]4 .
lJ
- I

c D = [~

~ [~
-4
E I
2 J F ~ [-4 :]
2
20 Chapter 1 Linear Equations and Matrices

",ul 0 ~ [~ ~ ~]
6. If possible. compute the indicated linear combination:
(a) C + £and£+C (b) A+B
(e) D- F (d ) -3C - 50
(e) 2C - 3£ (I) 2B + F
7. If possible. compute the indicated linear combination:
(a) 3D+2F (b ) 312A) and 6A
(e) 3A + 2A and 5A
(d ) 2(D + F) and 2D + 2F
(e) (2 +3) Dand2D+3D
(I) 3( B + D)
8. Ifpossible. compute the following:
(a) A T and (AT)T
(b) (C + £)T and CT + £T
(e) (2D + 3fY (d ) D _ D T
(e) 2AT + B (I) (3 D - 2F)T
9. If possible. compute the following:
(a) (2A) T (b) (A _ B )T
(e) (3 B T _ 2A) T (d) (3 A T _ 5B T ,r
(e) (_ A )T and _ (A T) (I) (C + E + F T)T
10. Is the matnx . [' 0]. 0 2 ..
a lmear combmatlOn of the matn - .
OJ
1 and ['0 OJ
0 ? Ju,tl.fy your answer.

.
II. Is the matnx [' '].
0 - 3 . . of the ma-
a hnear comblllatlOn

.
Inces
[, O] [' 0] .
0 1 and 0 0 ? Justify your answer.

12. Let

o
2
2 ,
2

If). is a real number. compute AlJ - A.

°
13. If A is an /I x /I matrix. what are the entries on the main
diagonal of A - AT? Justify )·our answer.
 1.3 Matrix Multiplication 21

Matrix Multiplication

The dot prod uCI, or inner product, of the II-vectors in R"

DEFINITION 1.6

[ ~;]
at]
ll2
a= . and b =
[
a" b"
is defined as
"
a· b = lllb t +a2b2 + . .. +all b., = L a,b, .·
,=1

The dot product of

EXAMPLE 1

is

u·, ~ (1)(2) + (-2)(3) + (3)(-2) + (4)(1) ~ - 6.

•
EXAMPLE 2 Let a ~ m ~mand b Ila b ~-U nd x

Solution
We have
a - b = 4x +2+6 =-4
4x+8 =-4
x = - 3.
•
11 0.."." , li»u ' ........... M",';. "

• , - I _. I ,,~ .'" - , .... - I>" I • • ,"' ,_•."-,..........

.. A ... . ....,.., .'I . . . . ~ " . ..... e _[.,, ' ...... >,
';, - ~ , . ".~,."., ... . ~

-t,. ....., " ~,~~- , ~,, ., '"

_ "' ,., '.". . . ' , ,t """'" ;.~ • .,.,."' -.. .,~ ",,,,,,-
... ..... ____ ~ ,'~ , ~ "~, M ·,I." _"" U. '"~ , H ,j' ." 'M I•
.; ........... ...
c-..., ~ ,(" . ~

•-' .,
, , ,
'" • , ,'.
~ ~,~

,~.-"

-,~ '

• -'
•
• • • '. '. '.
"
~
,t~
"

'
'"
.. 'j "
"
"
j
-~ "I "" ~ .... ' - '.
""",. u
,- "-,,,
<i h , ~ ",'''''
". ''" ,.,..... ,,"
_~""
. . . ...... ,..,'_a-""" .. . ,._ '-'.
"
. ...., , , " " _ ' "
- , ,~ ,
~'"'"

........". c•
,. ,, , :1 .-1'-; ,I -~ .
 1.3 Matrix Multiplication 23

A B AB
III Xp p X n m X"

FIGURE 1.5
ltd sizeofAB

Then
AB = [ (1)(- 2) + (2)(4) + (-1 )(2) (1 )(5) + (2)( -3) + (- 1)( 1) ]
(3)( - 2) + (1 )(4) + (4)(2) (3)(5) + (1)(- 3) + (4)( 1)

~~J
LeI
= [:
•
EXAMPLE S

h [~ -~ -n ond B ~ [; -~] .
- 2 2
Compute the (3 . 2) entry of A B.

Solution
If A B = C, then the (3. 2) entry of A B is C32, which is (row3(A» T , co12( B ). We
now have

[-n~-5 •
Let
EXAMPLE 6
3 ]
1
ond
B~ m
.
ItAB = [ 12
6,] findxand y .

Solution
We have

AB = [~ _~. 3] [!]~ [2 + 4X + 3Y] ~ [ I2] .

[ 4- 4+ y 6
y

Then

2 + 4x +3y = 12
y = 6.

so x = - 2 and y = 6.
•
24 Chapter 1 Linear Equations and Matrices

if A is anm x p matrix and B is a p x II matrix, then AB is anm x II

matrix. What about BA? Four different situations may occur:
I . BA may not be de fin ed; this will take place if /I t- III.
2. If BA is defined, which means that 11/ = fI, then BA is p x P while AB is
III x 11/; thus. if 11/ F p. AB and B A are of different sizes.

3. If A Band B A are both of the same size, they may be equal.

4. If AB and B A are both of the same size, they may be unequal.

If A is a2 x 3 matrix and B isa3 x 4 matrix.thenAB is a2 x 4 matrix while BA

EXAMPLE 7
is undefined. •

Let A be 2 x 3 and let B be 3 x 2. Then AB is 2 x 2 while BA is 3 x 3. •

EXAMPLE 8

Lei
EXAMPLE 9

Then
A~ [
- I
1
~] on" B= [~ :1
AB =[
-2
2
~] while BA = [ I
- I
7]
3 .

Thus AB =1= BA .
•
 1.3 Matrix Multiplication 25

LeI
EXAMPLE 11
A ~
[
-~
I
: '] and B ~
[
-2
3
3
2
4]
I .

T hen the second column of A B is

•
Remark If u and v are II-vectors (II x matrices), then it is easy to show by
matrix multiplication (Exercise 4 I) that

u. v = uTv.

T his observation is applied in Chapter 5 .

• The Matrix- Vector Product Written in Terms of Coiunms

LeI
(112

[ a" (l2J (1n a,,, ]

(12"
A ~

(l1~1 (1",2 {/~II'

26 Chapter 1 Linear Equations and Matrices

be an III x /I matrix and let

be an /I-vector, that is, an /I x I matrix. Since A is III X /I and c is /I X 1, the matrix

product Ac is the III x I matrix

(112
a,,,] ["] [(WW'(A »" ,c' ]
Ac =
[""{/~iI
{/21 a"
{/m2
(/ 2n

(11~"
C2

;"
(row2(A»T

= (row",(~»T . c
(2)
a"e, +a"" + ... + a,,,,,, ]
{/21CI + (122C2 + . .. + {/ln C"

[
{/"IICI + (1",2C2 + .. + {/",,,C,,
This last expression can be written as

CI
[a~'1::::] [:::]{/1~2
+C1
-
+ " ' +C
"
[::::]
{/I~" (3)

Thus the product Ac o f an III x 11 matrix A and an 'I x 1 matrix c can be written as
a linear combination of the columns of A, where the coetlicients are the entries in
the matrix c.

Let
EXAMPLE 12
A = [! - I
2
-3]
-2
and

Then the product Ac, written as a linear combination of the columns of A, is

Ac = [~ - I
2

If A is an III x p matrix and B is a p x n matrix, we can then conclude that

•
the Jth column of the product A B can be written as a linear combi nation of the
 1.3 Matrix Multiplication 27

columns of matrix A, where the coefficients arc the entries in the jl h column of
matrix B:

If A and B are the matrices defined in Exam ple [ L then

EXAMPLE 13

AB ~ [;
- I
!] [-~
5
3
2
4] - [: l~
I - 17 7
6]
l~ .

The columns o f AB as linear combinations of the columns o f A are given by

•
• Linear Systems

Consider the linear system of III equations in n unknowns,

aJlXj + {l12X! + .. . + {/lr,X n = hi

{/21Xj + {/nX2 + .. . + {/ 2" X n = b2
(4)

Now de fin e the following matrices:

(II !
[ U" U,,, ]

U [b'h2. ] .
{/2J (121 {/2"
A ~ x ~ b =

ami a,.," a,~", b m

28 Chapter 1 Linear Equations and Matrices

Then

::;:
Ax =
[ (I,~, I
a,"] [X']
a,"

(lmn
x,

J."
(5)
all \"1 + (l12X2 + .+ (l1"J." ]
all \"1 + (lnX2 + .+ (l2"J."

[
a",lxI + (1",2 tl + + (I"", J."
The entries in the product A x at the end of (5) are merely the left sides of the
equations in (4). Hence the linear system (4) can be wrillen in matri x form as

A x = b.

The matrix A is called the coefticient matrix o f the linear system (4), and the
matrix
(Ill al"
[ a"
(121 (122 ail, : h,
h2 ]

(I,:, ! (1m2 (I"", i h",

obtained by adjoining colu mn b to A, is called the augmented matrix of the linear
system (4). The augme nted matrix o f (4) is written as [A : b ]' Conversely, any
matrix with more than one column can be thought of as the augmented matrix
of a linear system. The coefficient and augmented matrices play key roles in our
method for solvi ng linear systems.
Recall from Section 1.1 that if

b l = b 2 = ... = bIll =0
in (4), the linear system is called a homogeneous system . A homogeneous system
can be written as
A x = O.

where A is the coefficient matrix.

Consider Ihe linear sys lem

EXAMPLE 14
-2x + z= 5
2x + 3)' - 4z = 7
3x + 2)' + 2;:: = 3.

Letting
 1.3 Matrix Multiplication 29

we can write the given linear system in matrix form as

Ax = b.
The coefficient matrix is A, and the augmented matrix is

[-~ 0
3
2
-4
2 n •
The maTrix
EXAMPLE 15
[~ ~l
- I 3
0 2 ,

is the augmented matrix o f the linear system

2x -y+ 3z= 4
3x + 2;: =
We can express (5) in another form, as follows, llsing (2) and (3):
5. •
a"x, + a" X.d .. . + a,. x. ]
{/2 1J.1 + an '\ 2 + .. + a2n \n
Ax =
[
{/"dXj + a,., 2_\2 + . + (I" ");. ,,

::::;: ] + [~;:;: ] + ... + [::~: ;: ]

[
{/", I Xj {/",2X2 am,:,x"

[~;: ] + .
= \1
[::::]
amJ
+ \~
{1m 2
+.x" [:::::]
(lmlt

= Xl col l(A) +x2 coh (A) + ... +X" coll/(A).

Thus Ax is a linear combination of the columns of A with coefficients that are the
entries of x. It [ollows that the matrix form of a linear system, Ax = b, can be
expressed as
XI COI l (A) + X2 COI2(A) + ... + XI! col,,(A) = b. (6)

Conversely, an equation of the form in (6) always describes a linear system of the
fonn in (4).

Consider the linear system Ax = b, where the coefficient matrix

EXAMPLE 16
30 Chapter 1 Linear Equations and Matrices

Writing Ax = b as a linear combination of the columns of A as in (6), we have

The expression for the linear system Ax = b as shown in (6), provides an

•
important way to think about solutions of linear systems.

Ax = b is consistent if and only if b can be expressed as a linear combination

of the columns of the matrix A.

Key Terms
Dot product (inner product) Coefficient matrix
Matrix~vector product Augmented matrix

.i' Exercises

L" '~ b ~[ -;llr'. b ~17.fiOd'.

III £Hm:isel' I alld 2, (;Ompllle a· b.
3.
I. (a) , ~[a b ~[ _~]
4. Determine the value of x so that v· w = O. where

(b ) ,~ [ =;J b ~ [_;]
(cJ
,~ ul b ~m 5. Detennine values of x and y so that v • w = 0 and v· u =

Idl , ~m b ~m Owb<re ' ~ [;] " ~Hl,"d U ~ m

6. Determine values of.{ and y so that v • w = 0 and v · u =

2. (a) ,~ [ ;J b ~m O.wb,re,~ [;] w ~ [-n,"d U~ [-H

(b ) '~ [ _ :l b ~[:] 7. Let w= [;~:: lcomputew. w.

(c) , ~m b ~n] 8. F;od ,II ,,'," of, '" ' '0< U·U~
50. wh,,, U~ [!1
(d ) ,~ [H b ~m 9. F;od,1I "',,,of, ,",b" ,., ~ ,. wh", , ~ [-1].
 1.3 Matrix Multiplication 31

10, lelA = [~ 2
-I 17. "'A ~ H n,"dB ~ [: -; :j
Compu te th e following entries of A8 :
IfA8 = [~]'findXand Y, (a) Ihe (1. 2) entry (b) the (2. 3) entry

CO/u;der t"e/ollowing matrices/or £\'l' l t';se,Y II t"mugh 15: C<) Ihe (3, I ) entry Cd) the (3. 3) c ntry

18. [I' f -, = 0 [I ~]andD= [-; -~l compute OIl

A ~ [; 2 3]
I 4 . and ' l l> .
19. u,
-I = [~ -I]
c ~ [;
n 2 -2]5 .
D- [3 A 4 .

Show Ihat A8/= BA.

~ [~ 2 ,,-=[-I~ ~2] .
-4 20. If A is Ihe matrix in Example 4 :md 0 is Ihe 3 x 2 matrix
I
E
n
11. If possible, co mpute the following:
(11111 e"cry onc of whose entries is zero, compute AD.
111 £urc:i,\'e,\' 2J (/1Il122 , 11'1

~]
- I
(a) A B (b) BA (e) F T[ 2
(d) C 8 + 0 (e) A8 + D 1. whcre 0 1
= OO -2
12. If possible, compUle the following:
(a) OA +8 (b) £C (e) CE all(/
Cd) £8 + F (e) FC + f) o - I
3 -3
13, If possible, compute the following:
2 5
(a) FD - 3B ( b) AB - 2D
2 1. Using the methexl in Example 11. co mpute the following
[e) F T8 +0 (d) 2F - 3( A£)
columns of A8 :
(e) 80 + A£ (a) thc first column (b) the third column
14. If po~ib l e. compute the following: 22. Using the methexl in Example II . compute the following
(a) A( BD) ( b ) ( AB ) I> co lumn.~o f AB :
(e) A (C+ E ) (d ) AC + A£ (a) the second column (b) the fourth column
(e) (2A8 )T and 2( A8 )T ( 0 A(C - 3 £ ) 23, Let
IS. If possible, compute the followin g:
(a) AT (b l ( AT)T
(e) (A8) T (d) B1' A T

[e) (C + £)1' 8 and C T B + £ TB Express A c as a linear combination of the columns of A.

(I) A (28 ) and 2( A8 ) 24. Lot

-;] [I -I]
16. Le I A = [1 2 -3 ],8 = [ - 1 4 2].and
-2
C = [ -3 0 I]. If possible. compute the following :
4 8 = ~ ~ .
(a) AB T (b) CA T (e) (HA T) C o -2
Cd ) A T 8 (e) CC T (0 CTC
Express th e columns of A 8 as linear combinations of the
(g) 8 /' CAA T columns of A.
32 Chapter 1 Linear Equations and Matrices

25. LetA=[~
-3
2 :l,"dB~
(, ) Verify lhal A B = 331 + 5a 2 + 2 3 ) , where
m 3j is the
33. Write the following linear system in matrix fonn:

hi
3x 2
+ 112 = 0
+ x] =O
jlh column of A for j = 1.2.3. hi - X2 = 0

. B = [(roW1(A))B] ,
34. Write the linear system whose augmented matri x is

n
(b) Venfyth:llA ( () .
row: A) B

U
1
26. (, ) Find a value of r so that A 8 T = 0, where (, ) - I 0
A = [r 1 - 2] and B = [I 3 - IJ. -,
,
3

[-~ n
(b) Gille an alternative way to write this product.
27. Find a value of r and a \lalue of ~' so that A S T = D. where - I 0
A=[l r l]andB=[ -2 2 .r]'
(b )
1 -, 3
0 0 0
35. How are the linear systems obtained in Exercise 34 re-
lated?
36. Write each of the following linear systems as a linear
combination of the columns of the coefficient matrix:
(a) 3xI + 21"1 + Xl - 4
XI - _11+4x; = -2
(b) - XI + Xz = 3
2xI - _I! = - 2
3xI + _I! = I
37. Write each of the following linear combinations of
columns as a linear system of the form in (4):

30. Consider the followinJ.! linear system:

211 + 3X2 - 3X1 + X~ + X5 = 7

3_l 1 + 2X1 + 3xs = - 2
38. Write each of the fo llowing as a linear system in ma trix
211+3x2 - 4X4 3 fonn:
l) + X4 + _IS = 5.

(a) Find the coefficient matrix.

(b) Write the linear system in matrix form.
(c) Find the augmented matrix.
31. Write the linear system whose augmented matrix is
39. Determine a solution to each of the following linear sys-
-2 - I o 4 tems. using the fact that Ax = h is consistent if and only

,:]
if h is a linear combination of the columns of A:
-3 2 7 8
1 o o 2
.
[ (, )
3 o 3 6

32. Write the following linear system in matrix fonn:

- 2xl + 3X2 = 5 (b)

XI - 5X2 = 4
40. Co",,,,,,,, e~ffi,;,"' m"ri, A '0 'h" , ~ [~] ;, ,
;olm;oo <0 'he ,y"em A, ~ b. wh,,, b ~ U1C,"
there be more than one such coefficient matrix? Explain.

43. jf A = [aii ] is an II x II matrix, then the Irace of A.

.
Tr(A). is defined as the sum of all elements on the main
diagolwl of A. Tr(A) = LUil Show each of the follow-
;=1
mg:
(a) Tr(cA) = c Tr(A). where c is a real number
(b) T r(A + B) = Tr(A) + Tr ( B )
(e) Tr(AB) = Tr(BA)
(d ) Tr(AT) = Tr(A)
(e) Tr(A T A) 2: 0
44. Compute the trace (see Exercise 43) of each of the fol-
lowing matrices:

(a) [~ ~J

Ie) [~ ~] !
45. Show Ihal there :Ire no 2 x 2 matrices A and B such that

AB - BA=[~ ~l
34 Chapter 1 Linear Equations and Matrices

Algebraic Properties of Matrix Operations

Theorem 1. 1 Properties of Matrix Addition

Lei A, B, and C be lit x /I matrices.
(a) A + B = B + A .
(b) A + (B + C ) ~ ( A + B )+ c.
(c) There is a unique lit x /I malri x 0 such Ihat
( I)

for any lit x /I matri x A. The matrix 0 is called the lit x /I zero matrix .
(d) For each lit x /I matrix A, there is a unique lit x n matrix D such that
A +D = O. (2)

We shall write D as - A, so (2) can be wri llen as

A + (- A) ~ o.
The matrix - A is called the negative of A. We also note that - A is (- I)A.
 1.4 Algebraic Properties of Matrix Operations 35

The 2 x 2 zero matrix is

o = [~ ~l
EXAMPLE 1 'f
A = [ 42 - 3
'] '

then

[42 -']3 + [00 0]o ~ [4+0 - ' + 0] ~ [ 4 -']

3 .

The 2 x 3 zero matrix is

2+0 3+ 0 2
•
o= [~ °o
EXAMPLE 2 •
Theorem 1.2 Properties of Matrix Multiplication
(a) If A, B. and C are matrices of the appropriate sizes. then
A( Be) ~ ( AB)e.
(b) Ir A, B. and C a rc mat rh.:cs uf tht: approp riate sizes. then

(A + 8 )C = AC + Be.
(c) Lf A , B , and C arc matrices of the appropriate sizes. then
C(A + B) = CA + C R. (3)
36 Chapter 1 Linear Equations and Matrices

•
Lc<
EXAMPLE 3

[~
- I

ond
A = [~ -3
2
!l B ~ 2
0
2
- I n
~ [~ ~]
0
-3
e 0

Then
3
-3 !l[~ 3 ~] ~ [43 30 5~
~ [~ -4 2 16
A( B e) ]
3 12

and
0
(AB )e ~ -I I~*
6
['9 - 8
16
0 ~] ~ [43 30 56]8 . •
-3
-8
0 3 12
16

LeI
EXAMPLE 4

Then
A = [~
-I n 3 -:1
2
B= [~
0
ond e~ U
-I~l
(A + BK = [~ 2
2
4l[~ ~] ~ [18 0]
I 3 - I 12 3
 1.4 Algebraic Properties of Matrix Operations 37

and (veri fy)

-I]~ [18 0]-

7 12 3
•

Theorem 1.3 Properties of Scalar Multiplication

If rand s are real n um~ rs and A and B arc matrices of the ap prop riate sizes, then
(a) r(sA) = (r.I")A
(b) (r+s)A = rA+.IA
(c) rCA + B) = rA + r B
(d) A(rB) = r(AB) = (rA)B

•
Let
EXAMPLE S
A = [~ -3
2
!] ond

Then
2(3A) = 2[ I ~ -9
6 9] [2412
12 =
12
- 18 2418] = 6A .

We also have

A(2B) = [~ - 3
2 - 10
o
16] ~ 2(A B',I.
26

Scalar mult iplication can be used to change the size of entries in a matrix to meet
•
EXAMPLE 6
prescribed properties. Let

Then for k = ~ ,the largest entry o f kA is I . A lso if the entries of A re present the
volume of products in gallons, for k = 4, kA gives the volume in quarts. •
38 Chapter 1 Linear Equations and Matrices

Theorem 1.4 Properties of Transpose

If r is a scalar and A and B are matrices of the approp riate sizes, then
(3) (AT)T = A
(b) (A+ B )T=A T +BT
(c) (AB)T=BTAT
(d) (rA)T = rAT

LeI
EXAMPLE 7
A _ [ 1 23 ]
- -2 0 1
and B ~[3 - I ']_
3 2-[

Then

""d B' ~ [ -~ ~]_

2 - I

Also,

Now

LeI
•
EXAMPLE 8

~] " d B ~ [~3 - ~]I _

Then
A B =[ 12
, -3
5] and (A B ) , [125 - ']3 .
=
 1.4 Algebraic Properties of Matrix Operations 39

On the othe r hand,

AT ~U -l] and
2
2

T hen

•
We also note two other peculiarities of matrix multiplication. If {/ and IJ are
real numbers, then ab = 0 can hold only if a or b is zero. However, this is
not true for matrices.

If

[ 4 -6]
EXAMPLE 9
A = [ 2I '4 ] ond B= -2 3 '

then nei ther A nor I:J is t he zero matri x. but A I:J = [~ ~ J. •

If a , b, and c are real numbers for which ab = lie and (J ",. 0, it follow s
that b = c. That is. we can cancel out the nonzero factor lI. However, the
cancellation law does not hold for matrices, as thc following example shows.

If

[-' 7]
EXAMPLE 10
ond c= 5 - I .

then

AB=AC =[,! l~l

but B i= c. •
40 Chapter 1 Linear Equations and Matrices

Key Terms
Properties of matrix :lddition Properties of scalar multiplication
Zero matrix Properties of transpose

-e-
Properties of matrix multiplication

Exercises
I. Prove Theorem 1.I(b). I I. Find two unequal 2 x 2 m:ltrices A and B such that
2. Prove Theorem 1.I (d).
3. Verify Theorem 1.2(a) for the following matrices:
All - [~ ~l
12. Find two different 2 x 2 matrices A such that A" = O.

[~
-I 3
A= B~ 13. Prove Theorem 1.3(a).
[ 1 - 3
14. Prove Theorem 1.3(b"1.

-a
15. Verify Theorem 1.3(b) for r 4. s - 2. and A
and
[~
4. Prove Theorem 1.2(b) and (c) 16. Prove Theorem 1.3(c).
5. Verify Theorem I.I(c) lor the following matrices: 17. Ver ify Theorem l.3(c) lor r = - 3.

[~ "~[~ [ ~ ;].
-3 1
A= - I -;] 3 and B~
-2 1
C = [ _~ -!]. 18. Prove Theorem 1.3(d·..
19. Verify Theorem 1.3(d) for the following m:ltrices:
6. Let A = [aij ] be the I! x matrix defined by (Iii = k
/I

and(lij =Oifi i=- j. Show that if B is any n x /I matrix. - I 3

B~
then AB = kB. [ 1 - 3
7. Let A be:lnm x II matrix :lnd C = [CJ C2 and r = - 3.

.
a I x II! matrix. Prove that

CA = L:>iAj.
j= l

whereA j is the jth row of A.

8 LetA=[ cosO sinO]

. - si n O cosO .
(a ) Determine a simple expression for A 2.
(b) Determine:l simple expression for A ].
(c) Conjecture the form of a si mple expression for A".
k a positive integer.
(d ) Prove or dispro\le your conjecture in part (c).
9. Find a pair of unequal 2 x 2 matrices A and B . other than
those given in Example 9. suc h that AB = O. 22. Determine a scalar r such th:lt Ax = rx . where
10. Find two different 2 x 2 matrices A such that

A2 = [~ ~l
 1.4 Algebraic Properties of Malrix Operations 41

23. Determi ne a scalar , such that All: = ' x. whe re (l.") For any seal:!r" show that 'X I is a solution.
(d) For any scalars r and J, show that ' XI + .I'X~ is a
solution.
37. Show that if All: = II has more than one solution, then
it has infi nitely many solu tions. (H illl: If Xl and X2 are
solut ions. l."onsidcr Xl = ' X I + J X2' where, + s = I.)
24. that if Ax = r x for" x 1/ matrix A. II X I mat ri x
PrO\'c 38. Show th:lt if X I and X2 are solutions to the linear system
x. and Sl:alar r, thcn Ay = ,y, where y = sx for any Ax = b. then X I - X2 is a solution to the associated ho-
scalar s. mogeneous system Ax = O.
25. l)etennine a Sl:alar .f such Ihal AZx = s x when Ax = r x .
26. Pro\'e 1llCOrem 1.4(a).
27. Prove Theorem I .4(h) and (d).
28. Verify Theorem 1.4(a), (b), ;llId (d) for

A = [~ 3 _;]. 8 = [ _~ 2
-']
5 .
fmd r = - 4.
29, Verify Theore m IA(c) for

3
B= [' -']
~ ~.

30. Le t

Compu te 8 r C and mult iply the result by A on the

ri ght. (Hi",: B TC is I x I ).
(c) Explai n why ( A8 T )C = ( 8 T C) A .
3 1. Determin e a constant k such thaI (k A )T (k A ) = I, where

A ~ [~:l '''h,re m""h,"o~ "'000" 'h" oo,'d

be used?
32, Find three 2 x 2 matri ces. A. B. and C such that AB =
ACwith B =FC and A =F O .
33, Le t A be an /I x /I matri x li nd c a real number. Show that
If cA =
O.then c= OorA = O .
34. Determine all 2 x 2 malrices A such Ihal AB = BA for
any 2 x 2 matrix B.
35. Show that (A _ 8 )T = A T _ 8 T.
36. Le t X I and X2 be solutions 10 tl:e homogeneous linear sys-
tem Ax = O.
(a ) Show that XI + X1 is a sol ution.
(b) Show that XI - X2 is a solu tion.
42 Chapter 1 Linear Equations and Matrices

III Special Types of Matrices and Partitioned

Matrices

An /I x /I matrix A = [(lij ] is called a diagonal matrix if (Ii) = 0 for i =1= j.

Th us, for a diagonal matrix, the terms off the main diagonal are all zero. Note
that U is a diagonal matrix. A scalar matrix is a diagonal matrix whose diagonal
clements are equal. The scalar matrix 1" = [dij ], where dl ; = I and dij = 0 for
i "1= j , is called the /I x /I identity matrix.

LeI
EXAMPLE 1

Then A, B, and l, are diagonal matrices; Band 13 are scalar matrices: and I) is
the 3 x 3 ide ntity matrix. •

It is easy to show (Exercise I ) that if A is any m x 1/ matrix, then

AI,, = A and I",A=A.

A lso, if A is a scalar matrix, then A = r 1" for some scalar r.

Suppose that A is a square matrix . We now define the powers of a matrix, for
p a positive integer, by

A P= A·A··· · ·A.
~
I' factor.

If A is II X II, we also define

Fm no nneg>lTive inTegers p ,inri (/ , The f>lmili>lr law, of exponenTs fm The real num-
bers can also be proved for matrix multiplication of a square matrix A (Exercise 8):

It should also be noted [hat the rule

(ABV = AI'BI'

docs not hold for square matrices unless A B = B A. (Exercise 9).

 1.5 Special Types of Matrices and Parti tioned Matrices 43

An /I X /I matri x A = [aU ] is called upper triangular if aU = 0 fo r i > j.

It is called lower triangular if a ij = 0 for i < j. A diagonal matrix is both upper
tri angular and lower triangular.

The malrix
EXAMPLE 2

is upper rriilnglllnr, anti

is lower triangular. •

A matri x A with real entries is called symmetric if A T = A.

DEFINITION 1.8

A matrix A with real e ntries is called s kew symmetric if A T = - A.

DEFINITION 1.9

Un
2
EXAMPLE 3
A ~ 4
S
is a symmetric matri x.
•
EXAMPLE 4
B ~ [-~ -n, ,
- 3
2
0
4
' kew 'ymmell'e mo,,'x. •
if A is symmetric or skew ~y m­
metric, then A is a square matrix. If A is a symmetric matrix, then the entries of A
are symmetric with respect to the main diagonal of A. A lso, A is symmetric if and
only if a ;j = a j; , and A is skew symmetric if and only if aij = - a j; . Moreover, if
A is skew symmetric, then the entries on the main diagonal of A are all zero. An
important property of symmetric and skew symmetric matrices is the following: If
A is an /I x /I matrix, then we can show that A = S + K , where S is symmetric
and K is skew symmetric. Moreover, this decomposition is unique (Exercise 29) .

• Partition ed Matrices
If we start Ollt with an 11/ x /I matrix A = [a j) ] and then cross out some, blll not
aiL of its rows or columns. we obtain a submatrix of A.

Lei
EXAMPLE 5
~]
2 3
4 - 3 .
o S -3
44 Chapter 1 Linear Equations and Matrices

If we cross out the second row and third column, we get the submatrix

2
o
•
A matrix can be panitioned into submatrices by drawing hori zontal lines be-
tween rows and vertical lines between columns. Of course, the partitioning can be
carried out in many different ways.
The mati ix
EXAMPLE 6

can be partitioned as indicated previously. We could also write

;:;:c~;: ] ~"
(/12 i(/IJ
A ~ [a" __ an~?3_~_a23 (134
(/21

(/31
(/4 1 a42
i a33
: {/4J
i A" (115
(144 : (145
[ AI2 ~,, ]
A22 A"
(I)

which gives another partitioning of A. We thus speak of partitioned matrices . •

The augmented matrix (defined in Section 1. 3) of a linear system is a partitioned

EXAMPLE 7
matrix. Thus, if Ax = b, we can write the augmented matrix of this system as
[A i b] •
If A and B are both 11/ x /I matrices that are pilnitioned in the same way, then
A + B is produced simply by adding the corresponding submatrices of A and B.
Similarly, if A is a panitioned matrix, then thc scalar multiple cA is obtained by
forming the scalar multiple of each submatrix.
If A is partitioned as shown in (I) and

I
b 11 bn i bD b l4
b21 bn i 1m b",
B ~ b JI b J2 i b JJ b J4 B"
B2\ B" ]
:~: .
b 41 b 42 : b43 b" _B31
----------f--
b 5J b 52 : b53 b"

then by straightforward computations we can show that

(All B\ I + AI2B21 +
A\3BJ1 ) (A l1 B I2 .41 2B22 + +
ADBJ2)]
AB = -------------------------------- - - - - - - - - - - - - - - - -----------------
[ (AIIB I I I AnBl! I AnB31) (Ali BI l -1 AnBll I AnB3l)
 1.5 Special Types of Matrices and Parti tioned Matrices 45

Lei
EXAMPLE 8

A" ]
An

and let

B ~ [--f--- -l-----~--H- - -~- -~i]

- 3 - I 2: I 0 - 1
B12]
Bn .

Then

where ell should be AII BI I + A1 2B21. We verify thai ell is this expression as
follows:

AIIBII +A12 B21 = [~ m~

0
~J + [~ -~][ -~
3
- I ~]
~ [~ ~] + [~ -~]
0 3
2 10

~ [! ~]
3
= Cli.
12
•
This method of multiplying partitioned matrices is also known as block mul·
tiplication.

Lei
EXAMPLE 9
- I 0]. a nd D = [:
8
7
-4]
5 .
46 Chapter 1 Linear Equations and Matrices

Then we have

[B 9
6
8
7
-4] 5 .

8
7
- I
•

the augmented matri x of the linear

system Ax = b is a partitioned matrix.

• Nonsinguiar M atrices

An /I X /I m>ltrix A is c>llle(! nonsinglll:-.r, or invertih!e, if there eXiST;\; >In /I x /I

DEFINITION 1. 10
matrix B such that A B = B A = In; such a B is called an inverse of A. Otherwise,
A is called singular. or noninvertible.

Remark In Theorem 2.1 I, Section 2.3. we show that if AB = In, then BA = In.
Thus, to verify that B is an inverse of A. we need verify only that AB = In.

EXAMPLE 10
LetA=
[23]
2 2 andB = [-I ']
I _~ .SinceAB = BA = /2 ,weconcludcthat
B is an inverse of A.
•
Thearem 1.5 The inverse of a matrix. if it exists, is unique.
 1.5 Special Types of Matrices and Parti tioned Matrices 47

Because o f this uniqueness, we write the inverse of a nonsingular matrix A as

A-I. Thus

LeI
EXAMPLE 11
A = [~ ~].
If A -1 exists, let

Then we must have

'] [a, ~J = h = [~ ~].

4

so that
a+2,
[ 3a+4c
"+2d]
3b+4d =
['
0
0]
I .
Equating corresponding entries of these two matrices, we obtain the linear systems

a+2c = I b+2d = O
"nd
3a+4c = O 3b+4d = 1.

The solutions are (verify) a = - 2. c = ~,b = I, and d = - !. Moreover, since

the matrix
a
[c
,,] ~[-; :]
d 1:-'2
also satis ties the properly that

[-, - 4,][, ,] ~ [, 0].

~ 3 4 0 I .

we conclude that A is Ilonsingular and that

A
_1 _[-2 ,]
- J J'

LeI
:2 -'1
•
EXAMPLE 12

I f A-I exists, lei

Then we must have

AA-I ~ [2' '][a

4cd b] ~ J2 ~ ['0 0]
[ ,
48 Chapter 1 Linear Equations and Matrices

so that
0+2, "+2dl ~[ 1 OJ.
[ 2o+4c 2h+4d 0 I

Equating corresponding entries of these two matrices, we obtain the linear systems

a + 2c = I h+2d = O
and
2a+4c = O 2h+4t1 = 1.

These linear sysu::ms ha ve no solutions, so our a~sumplion Ihal A-I exisls is in-
correct. Thus A is singular. •

Theorem 1.6 If A and B are both nonsingu lar /I x /I matrices, then A B is nonsingular and
(A B )-I = B -IA- I .

. •

Corollary 1. 1 If A I. A2 . . ... A, arc II x /I nonsingular matrices, then A I AI' .. A r is nonsingular

and (A I A 2 '" A,)-I = A;I A ;~I . .. Al l .

Theorem 1.7 If A is a nonsingu lar matri x. then A -I is nonsingular and (A -1)-1 = A.

Theorem 1.8 If A is a nonsingu lar matrix, then AT is nonsingular and (A -ll = (AT)-I .

If
EXAMPLE 13
 1.5 Special Types of Matrices and Parti tioned Matrices 49

then from Exam ple II

A
_, _[-2 ,]
- 3 1
2" -2:

Also (verify),

and

Suppose that A is nonsingular. Then A B = AC implies that B C (Exer-

•
cise 50), and AB = 0 implies that B = 0 (Exercise 51).
It follows from Theorem 1.8 that if A is a symmetric nonsinguiar matrix, then
A -1 is symmetric. (See Exercise 54.)

• Linear Systems and Inverses

If A is an matrix, then the linear system Ax = b is a system o f 1/ equations in
/I x 1/
/I IInknown .~ _ Sllrro.~e
111011 A is nomingll l>lT. The n A -I exists, ami we en n multiply
Ax = b by A-Ion the left on both sides. yielding
A-1(Ax) = A-1 b
(A-1A)x = A-1b
J"x = A-Ib
x = A-1b. (2)

Moreover, x = A-I b is c learly a solution to the given linear system. Thus. if A is

nonsingular. we have a unique solution. We restate this result for emphasis:

If A is an 1/ x 1/ matrix, then the linear system Ax = h has the uniq ue solution

x = A-l b. Moreover. if b = 0, then the uniq ue solution to the homogeneous
systemA x = O is x = O.

Suppose that A is the matrix of Example II so that

EXAMPLE 14

A
_, [-2- t,] .
= ~
If
52 Chapter 1 Linear Equations and Matrices

Key Terms
Diagonal ma tri x Symmetric matrix Nonsingul:lr (invertible) matrix
Identity matrix Skew symmetric matrix Inverse
Powers of a matrix Submatrix Singular (noninvertible) matrix
Upper triangular matrix Partitioning Properties of nomingular matrices
Lower triangular matrix Partitioned matrix Line:lr system with nonsingular coefficient matrix

_ , . Exercises

I. (a) Show th:lt if A is any III )( n matrix. then I", A = A 12. For a nonsingul:lr matrix A and a nonneg:ltive integer p .
and A I" = A. show that ( A P) - I = (A- I)".
(b) Show that if A is an II x I! scalar matrix. then 13. For a nonsingular matrix A :lnd nonzero scabr k. show
A = r I. for some real number r. that (kA) - 1 = t A- 1.
2. Prove that the sum. product, and scalar multiple of diag. 14. (a) Show that every sC:llar matrix is symmetric.
onal. scabr. and upper (lower) tri,lIlgular matrices is di· (b) Is every scalar ma trix nonsingular? Expbin.
agonal. scalar, and upper (lower) triangular, respectively. (e) Is every diagonallll:ltrix a scalar matrix? Exp bin.
3, Prove: If A and 8 are I! x II diagonal matrices. then 15. Find a 2 x 2 matrix 8 f- o and B f- I! such that
AB = BA.
4. Let AS= BA . where A = [ '2 ']
1 . How m:lny such matn.·
ces S are there?
-3
2
16. Find a 2 x 2 matrix B f- o and B f- 12 such that
o AB=BA. Where A= [~ ']
1 . How many such matn.·

Verify that A + 8 and A B are upper triangu lar. ces B are there?
5. Describe:lll matrices that are both upper :lnd lower trian· 17. Prove or disprove: For any 11 XII matrix A. A T A = AA T.
gular. 18. (a) Show tlwt A is symmetric if and only if (Ii) = {I i i
furalli.j.
6. LetA=[~ ~~]and S =[~ -~lcomputeeach (b) Show that A is skew symmetric if and only if a ij =
of the following: - a ii foralli.j.

(, ) A' (b ) 8 3 (c) ( AS )! (e) Show that if A is skew symmetric. then the elements
on the main diagonal of A are all zero.

[i H :]
0 0
7. Lo< A ~
-l] and S
0
19. Show that if A is a symmetric matrix. then A T is sym-
metric.
Compute e:lch of the following: 20. Describe all skew syr.lmetric scalar m:l trices.
(a) A 3 (b ) S ! «) (A 8 )3 21. Show that if A is any III x n matrix. then AA T and A T A
alC SY IllIllCtllC.
8. Let p and q be nonnegative integers and let A be:l square
matrix. Show th:lt 22. Show that if A is any I! x I! matrix . then
(a ) A + A T is symmetric.
(b) A - A T is skew symmetric.
9. If AS = BA and p is a nonnegative integer. show that 23. Show that if A is a symmetric m:ltrix, then A I, k
{AB )P = A PB ". 2.3 ..... is symmetric.
10. If p is a nonneg:ltive integer and e is a scalar. show that 24. Let A and S be symmetric m:ltrices.
(eA)!' =e PA ". (a) Show that A + B is symmetric.
II. For:l square ma tri x A and:l nonnegative integer p. show (b) Show that AS i. symmetric if and only if AS
[hat ( A T)" = (AI,)T. SA .
 1.5 Special Types of Matrices and Pa rtitioned Matrices 53

25. (a) Show that ir A is an upper triangular matrix. then Find the solutio n x.
AT is lower triangular. 38. The linear system A ~ .~ = b is such that A is nonsingular
(b) Show that if A is a lower triangular matrix. then AT wi lh
is upper lriangul:lr.
26. If A is a skew symmetric m.atrix. whal Iype of malrix is
AT? Justify your answer. Find the solution x.
27. Show that if A is skew sym m~t ric, then the elements on 39. The linear system AT x = h is such that A is nonsingular
lhe main dia gonal of A are all 1.ero. wit h
28. Show that if A is skew symllletric, the n A' is skew sy m·
metri c for any positive odd inlCger k. Al= [~ ~] and b =[ _~] .
29. Show 1hat if A is an It x II ma\Jix. then A = S + K . where Find the solution x.
S is sy mmetric and K is skew sy mmetric. A lso show that
40. The linear system C T Ax = b is such that A and C are
this decomposition is unique. (Hilll : Use Exercise 22.)
nonsingular. wi th
30. Let

: -n·
Find the matrices Sand K desc ribed in Exercise 29.
Find the solution x.

31. Show that the m:l1rix A = [! !] is singular.

41. Consider th e linear syMem A x = h. where A is the mao
trix defined in Exercise 33(a).

o Finda sOlutiOnif b =[~].

[~
(a)
32. IfD = -2
o . . .'b [5]
( b ) Fmd a solutIOn 11 = 6 .

n
33. Find the inverse of each of the following matrices:
42. Find t...."O 2 x 2 singula r matrices whose sum is nonsin·
(a) A = [! ;] (b) A = [~ gular.

34. If A is a nonsingul ar matrix whose inverse is [~ :l 43. Find twO 2 x 2 nonsUlgular matrices whose sum ii sin·
gular.
44. Pro\'e Corollary I. L
fi nd A.
35. If 45. Pro\'e Theorem 1.7.
46. Prove Ihal if one row (column) o f the n X II matrix A con·
and 1 sists e nti rely of zeros. lhen A is singular. ( Hinl : Assume
B- -- [ 3'
lhal A is nonsingular; that is, th ere exists an /I x /I matrix
B such lhm AB = BA = I". E~labli s h aconlradiclion.)
fi nd (AB )- I.
47. Prove: If A is a diagona l illlitrix with nonzero di·
36. Suppose that
agonal el11ries {/11.{/ll ••••• II" • • then A is nonsingu·
A-
I
=[: ~l lar and A- I is a dillgonal malrix Wilh diagonal en tries
1 / 11 1 1. l / lIll ..... 1/,,"".
Solve the linear system Ax = h for each of the following
o
matrices b:
48. Lo< A = [~ -3
o
49. For an /I x /I diagonal matrix A whose diagonal entries
37. The linear sys te m AC x II is such that A and Care
arc lIll' li n . .... a,,", compute AI' for a nonnegative inte·
nonsi ngul ar with
ge r fJ.
50. Show Ihat if A B AC and A is nonsin~ular. then
1J =c.
54 Chapter 1 li near Equations and Matrices

5 I. Show that if A is nonsingular and A H =0 for an /I x /I

nlatrix H. then H = O.

52. Let A = [:. : l Show that A i~ nonsingular if and

only if lid - be 1= o.
53. Consider the homogeneous sys tem Ax = O. where A is
/I X 11.If A is nonsingular. ~how th llt thc only solution is
the trivial onc. x = O.
54. Pro\·c that if A is symmetri c and non ~ ingular. then A- I
IS symmetric.

55. Formulate the methoo for adding panitioned matrices.

9nd verify your methoo by partitioning the matrices

A= [i -3
3

1I11WO different ways and finding their ~u m .

56. Let A and lJ be the follow ing matrices:

AJ~
1 3 4
2 3 - I
3 2 1
-I 3 2

II - I
2
3
4
5
1]
>eo

-:]
2 3 4

8 = [j 5
1
2
3
4
3
4
2
2
5
6
3
7
1
.

Find A B by partitioning A and B in twO different ways.

57. What type of matrix is a linear combi nation o f symmetri c
matrices? Justify your answer
58. Whm type of matrix is a linear combination o f scalar ma-
trices? Justify you r answer.

:
 55

Matrix Transformations

R2 denotes the set of all 2-vectors and RJ denotes the set of all
3-vcctors. It is convenient 10 represent the clements of R2 and R3 geometrically
as directed line segments in a rectangular coordinate system.·

The veclOr

x ~ [~.l
in R2 is represented by the directed li ne segment shown in Figure 1.6. The veclOr
56 Chapter 1 Linear Equations and Matrices

z-axis
y-axis

y (x. y)

/ (X_y_ Z)

- - - - -;;f- - + - -_ x_axis
-::::/tco~=---+---:;?)c,-- )'-axis
--x-_____
o
x-axis

FIGURE 1.6 FIGURE 1.7

in R3 is represented by the directed line segment shown in Figure 1.7.

Fig ure I.R shows ef:Om~l ri c represent>llions of the 2-vcclors

EXAMPLE 1

in a 2-dimcnsional rectangular coordinate system. Fi gure 1.9 shows geometric

representatio ns of the 3-vectors

in a 3-dimensional rectangular coordinate system.

•
)'

,
"
., ., .,
-2 0
, a )'

"

FIGURE 1.8 FIGURE 1.9

 1.6 Matrix Transformations 57

If A is an III x II matrix and u is an II-vector, then the matrix product Ji u is

an III -vector. A functi on I mapping R" into R'" is denoted by I: R" -'jo R"' .' A
matrix transformation is a function I: R" -'jo R'" defi ned by I(u) = Au. The
vector I(u ) in R'" is called the image of 1I , and the set o f all images o f the vectors
in R" is called the ra nge o f I .

EXAMPLE 2
(a) Let I be the matrix transformation defi ned by

f(U)~[~ ~]U
The imageof ll = [ -~J is

.
and the Image 0 1 2 .[,] . ['0] ..
TS 5 (venly).

(h) LetA = I'

[ _ 2I
0] . . .
I . and consIder the matnx transforrnatlOn defi ned by

I(u) = Au .

Th," 'he 'm"ge o f m"m, 'he 'moge ul m.,m""" 'he 'm"ge 01

[-n"[~] ("rify) •

Observe that if A is an III x /I matrix and f: R" -'jo R'" is a matrix transfor-
mation mapping R" into R'" that is defi ned by I(u) = Au, then a vector w in R'"
is in the range of I onl y if we can find a vector v in R" such that f( v ) = w.
58 Chapter 1 Linear Equations and Matrices

EXAMPLE 3
Let A = [ - 2I '3] and consIder
. . transfonnatlOn
the matnx . defined by feu) = Au.
Determine if the vector w = [ _ ~] is in the range of f.

Solution
The question is equi valent to asking whether there is a vector v = [ :' ] such that
f( v ) = w. We have 2

Av -- [ - 2vl
" +"'] -w-[ 4]
+3V2 - - - I

VI + 2V2 = 4
- 2vl + 3V2 = - 1.
Solving this linear system of equations by the familiar method of elimination. we
get VI = 2 and V2 = I (verify). Thus w is in the range of f. In particular, if

v= [~llhen f( v) = w. •

.
LeI f: R2 -+ R2 be the matrix transfonnation de fined by
EXAMPLE 4

(x. )')

" Thus. if u = [::],then

. o¥ :------ ,
f lu)
(x. - yJ I(u ) ~I ([;,]) ~ [ _;] .
FIGURE 1. 10 Reflection The effect of the matrix transformation f. called refl ection with respect to th e
wilh respecllo (he x-axis. x-axis in R 2, is shown in Fi gure 1.10. In Exercise 2 we consider reflection with
respect to the y-axis . •
Let f: R3 -+ R2 be the matrix transfomlation defined by
EXAMPLE S

Then

I(u) ~I ([n) ~ [;.].

 1.6 Matrix Transformations 59

(x.y.:) L
"
o y

FIGURE 1.11 x

Figure 1.11 shows the effect of this matrix transformation, which is called projec-
tion into the xy-plane. (Warning: Carefully note the axes in Figure 1.11 .)
Observe that if

;;;;=d----.~ )" where.l" is any scalar, then

Projection I(v) ~ [ ; . ] ~ I( u).

FIGURE 1.12 Hence, infinitely many 3-veetors have the same image vector. See Fi gure 1.12.

No" thm the im'ge of the 3-vwm v [~] ""de< the m,,,ix """,fotm"ion
g: R J -+ R J defi ned by

i, [~l The elTw of thi' mmrix t"n,fonnmion i, , hown in Fig"" 1. 13.

•
Lei f: RJ -+ R·1 be Ihe matrix transfonnation defi ned by
EXAMPLE 6

' 0 0]
fe u ) =
[DO,
a r a lI .

where r is a real number. It is easil y seen that fl u ) = r u. If r > I, f is called

dilation ; ira < r < L f is called contraction. Figure 1.14(a) shows Ihe vector
60 Chapter 1 Linear Equations and Matrices

~
(""'d
o Y ko - -Y
, (x.)' 0)

FIGURE 1.13

II(U) = 2u, and Fi gure 1.14(b) shows the vector 12(u) = ~ u . Thus di lation
stretches a vector, and contraction shrinks it. Similarly, we can define the matrix
transformation g: R2 -)0 R2 by

We also have g(u) = r u, so again if r > I. g is called dilation: if 0 < r < [, g is

<:ulled cun tnll:tiull. •

oy-=- - - - - - -y

FIGURE 1. 14 (a) Dilation: r> I (b) Co~tmction:O<r< J

 1.6 Matrix Transformations 61

Suppose that we rotate every poinl in R2 counterclockwise through an angle ¢

EXAMPLE 8
about thc origin of a rectangular coordinate system. Thus, if thc po inl P has
coordinates (x. y), then aner rotating, we gCllhc point p i with coordi nates (x ', ),').
To obtain a relationship between the coordinates of p i and those o f P, we let u be
the vector [: l which is represented by thc directed line segment from the origin
to P (x . y). See Figure 1.15. Also. leI () be the angle made by II with the positive
x-axis.

,.
P'(x' . y')

ft" )
\ P(x . y)

"
-,rl''''''-'------ "
FIGURE 1. 15 Rotation

Letting r denote thc length of Ihc di rected line segment from 0 10 P , we see
rrom Figure I. J 5(a) that

x = rcos f}. y= r sin f} (I)

on"

X' = r cos(f} + ¢) . y' = r sin(f} +¢). (2)

By the formulas for the sine and cosine of a sum of angles, the equatio ns in (2)
become

X' = r cos f} cos¢ - r sin e si n ¢

.v' = r sin f}cos¢, + r cose sin¢' .

Substitllling the expression in (1) into the last pair of equations, we obtain

x '= xcos¢'-ysin¢' . y' = xsin¢, + ycos¢' . (3)

Solving (3) for x and y . we have

x = x'cos¢+y'sin ¢ and y=-x ' sin ¢+y'cosl/J . (4)

62 Chapter 1 Linear Equations and Matrices

Equation (3) gives the coordinates o f p i in tenns of those of p, and (4) expresses
the coordinates of P in tenns o f those of p ' .

We may also pcrfonn this change of coordinates by considering the matrix

transformation [: R2 -+ R2 defi ned by

(5)

Then (5) can be written, usi ng (3), as

It then follows that the vector fe u ) is represented by the directed line segment
from 0 to the point p i Thus, rotation counterclockwise through an angle c/J is a
matrix transformation. •

Key Terms
Matrix transfonnation Image Dilation
Mapping (function) Reflection Contraction
Range Projection Rotation

.,.W Exercises

III £Tercises I Ihmugh 8, skl'fch u alld ils image IIlIder each 4. r: R 2 --;. R" is a counterclockwise rotation through ~ rr
girer: malrix lrall.iformalioll f.
radians; u = [ =~]
5. f: R2 _ R2 defined by

2. f: R 2 -+ R2 (rcHcction with rCS llcct to the y-axi~) de-

fined by
! ([X])~[
y
- 0l 0][,].
- I )" .
" ~[3]
2

! ([;.])~[-~ m;l " ~[ -;] 6. f: R2 _ R 2 defined by

3. r: R2 _ R2 is a counterclockwise rotation through 30":

" ~[ -;J
, ,'_ .'-..,
'([;11-[: :] [;1 .. [:]
, ...' ..'- .
~,_ _
11:11·1; ::111 .. 1;1
. ""... " , "O f , ' _ , '., .. _ "
_;c__ _ ,., , .. , _,. .....

" H] ,•. []
" .. [;]

"
::::1:·