MopULE Matrices

1.1. INTRODUCTION

A matrix is defined as a rectangular array (or arrangement in rows or columns) of scalars

subject to certain rules of operations.
If mn numbers (real or complex) or functions are arranged in the form of a rectangular
array Ahaving m rows (horizontal lines) andn columns (vertical lines) then A is called an
m X n matrix. Each of the mn numbers is called an element of the matrix.
There are different notations of enclosing theelements constituting a matrix viz. (], ).
I|||but square bracket is generally used. An m xn matrix is also called a matrix of order m x n.
2 -1 51
For example, 3 0 4
is a 2 x 3matrix or matrix of order 2 x 3. It has two rows and
three columns. The numbers 2, -1, 5, 3, 0, 4 are its elements.
a13 ... din

a21 L2 23 ....d2n
An m x n matrix is usually written as a¡1 g2 g3 ....dgn

ml am2 am3 ..amn

Here each element has two suffixes. The first suffix indicates the row and the second
suffix indicates the column in which the element lies. Thus, aoa is the element lying in the
second row and third column,a,, is the element lying in the ith row and jth column.
For brevity,a matrix is usually denoted by a single capital letter A, B or Cetc.
Thus, an m X n matrix A may be written as
A=[a,]m xn or A= [a,], where i= 1, 2, 3, .., mij=1, 2, 3, ... n
1.2. APPLICATIONS OF MATRICES

In Algebra, the matrices have their largest application in the theory of simultaneous equations
and linear transformations. e.g., the set of simultaneous equations,

may be symbolically represented by the equation,

AX = B

1
 ENGINEERING
2 ATEXTBOOK
OF MATHEMATICS

a1 12 a13 ,B= b
where A= 1 dg2 dg3 X=
ba highex
ag1 ag2 dg3
matrices has been found of great utility in many branches of
The theory of differential equations, astronomy, mechanics, theory of
Ihathematics such as algebraic and
electrical circuits, quantum mechanics, nuclear physics and aerodynamiCs.
are
a determinant and a matrix but the two
Note. There is a similarity in the way of writing no numerical value. Matrix is
entirely different. A
determinant has a numerical value but a matrix has
only a convenient way of storing information.
1.3. TYPES OF MATRICES
numbers.
(1) Real Matrix, A matrix is said to be real if all its elements are real
W5 -3 1 is a real matrix.
e.g.,
0 -V2 7
(2) Square Matrix. A matrix in which the number of rows is equal to the number of
columns is called a square matrix, otherwise, it is said to be a rectangular matrix.
Thus, a matrix A= [a.]. v is a square matrix if m=n and a rectangular matrix if m=n.
A square matrix having n rows and n columns is called "a square matrix of ordern' or
"an n-rowed square matrix",
d12
e.g., d1 is a 3-rowed square matrix.
23
g1 32 a33.
The elements a,1s Ag9s l¡3 Of a square matrix are called its diagonal
elements and the
diagonal along which these elements lie is called the principal or leading diagonal.
In a square matrix A = (a,],
() for elements along the principal diagonal, i=j
(ii) for elements above the principal diagonal, i<j
(ii) for elements below the principal diagonal, i>j
(iv) for non-diagonal elements,
The sum of the diagonal elements of a
square
i#)
matrix is called its trace or spur. Thus,
trace of the n rowed square matrix A =la,| 1s a
tag2 t agg t ... ta,n
(3) Row Matrix. A matr1x
oteiy of order 1 X n is called having only one row and any number L=1
of columns L. e.,
a row matrix, e.g. (2 5
()Column Matrix. A matrix -3 0] is a row
having only one column and any [Link]
number rows i.e., a
matrix of order m x 1 is called a column matrix. e.g., 0 is a
column matrix.
Nell Matrix. A matrix in which
each element is zerois called a
matrix or a zero matrix. Anull matrix of
order mxn is denoted null matrix orvoid
by Om xn'
MATRICES 3

o o=0,2:00 0 0 o =0,x4
0 0
(6) Sub-matrix. Amatrix obtained from a given matrix Aby deleting some of its rows
or columns or both is called a sub-matrix of A.
[0 -1 2 5
Thus, B= [3 01 is a sub-matrix of A=3 5 0 7
1 4
1 6 4 - 2
obtained by deleting the first row, second and fourth columns of A.
(7) Diagonal Matrix. Asquare matrix in which all non-diagonal elements are zero is
called a diagonal matrix.
Thus, A= [a,], xyis a diagonal matrix if a,, =0for izj.
2 0

For example, A=|0 1 0 is a diagonal matrix.

0 0
An n-rowed diagonal matrix is briefly written as diag. [d,, d,, ..., d,], where
d,, d..... d, are the diagonal elements. Thus, the above diagonal matrix Acan be written as
diag. [2,- 1, 0].
(8) Scalar Matrix. A diagonal matrix in which all the diagonal elements are equal to a
scalar, say k, is called a scalar matrix.
Thus a scalar matrix is a square matrix in which all non-diagonal elements are zero
and all diagonal elements are equal to a scalar, say k.
0 when i #j
i.e., A= [al. v, is a scalar matrx if a, 1k when i=j
0
[2 0 07-50 -50
0

For example, 0 2 0 0 0
are scalar matrices.
0 0 -5
0 0 2]| 0 0 0 -5
(9) Unit Matrix or Identity Matrix, A scalar matrix in which each diagonal element
1s unity (i.e., 1)is called a unit matrix or identity matrix.
Thus, a unit matrix is a square matrix in which all non-diagonal elements are zero and
all diagonal elements are equal to 1.
(0 when i# j
i.e., A= [a),xn is a unit matrix if a,, =j1 wheni=j
A unit matrix of order n is denoted by I,. If the order is evident, it may be simply
denoted by I.
1 0 07
Thus, 0 1 0=1,
0 0 1
(10) Upper Triangular Matrix. A square matrix in which allthe elements below the
principal diagonal are zero is called an upper triangular matrix.
 or matrix if if
elementsingular entries
it ele matrix
the
MATHEMATICS
the B)
define
above below its
= all
triangular
A single non-zero
can
either as
ENGINEERING elements equal.
(written
a
be
to we
if
Thus,
matriy
elements called said words.|>1
having
mnatrix. the
a are j.
Thus, equal if and i-)
1.
stochastic
i>). matrix. elements
only is
is
A |A=0 other is
triangular all i<j. i elementmatrix matrix | matrix. row
for which
matrix. [Link] andall
TEXTBOOK
OF triangular all In for each
0
for for
to corresponding since =0 tridiagonal a
diagonal.
=
a,,
0 which said ifB b,, one square
triangular
the
a,
called
in
=
in triangular
if
matrix upper matrix
ifa,, =
are A a,
= onlymatrix. matrix is if elements
Matrices.
A matrix tridiagonal 1s
A matrix lower in B
then having super A
a
triangular an square
A
triangular a
matrix and
called
(ii)
theirB=[b,),pxoelement
a triangular.
singular
Tidiagonal
and
is
matr1X
the
1s lower is A matrix
square matrices
and Non-singular sub-diagonal
be square
of
3 Matrix. is |0. a to sum
5
zero A single 1s L23 g3 43
upper a
called A lower
order Matrix. 41
-3
2 41-1
0 Matrix. said 0
A the
lower are Two and
Matrix. a
A -
-3 is
12 22 32 Matrix.
3 1
Triangular0 same q is | and
6 2
diagonal if [a,]
diagonal,
1s
an is a or Matrices.
[a,]mxn
=
Element
3 non-singular
=2 0
zero is triangular n
the =pandmatrix Singular
(15)
and =
Tridiagonal =| non-negat
=[a,),xn
2 x, -1 5 3
Triangular A
A A Stochastic
are la,), have matrix example,
example.
For diagonal example,
For principal = example, leading
Lower Equal A m Singleany
= upper theyif Thus and real
A A Thus,
Thus, Thus (12)the (13)if (14) =0 For (16)the For (172 are
(11)principal either only (0) matrix. |A| in A ments
above only
and as:
is
4
 5
MATRICES

aj =1for i= 1, 2,
A=[a,], x, Will be a stochastic matrix if a,, > 0for 0siSn, 0 sjsnand j=1

n. e.g., A=
[2/3 V/3 is a stochastic matrix.
1/4 3/4

1.4. ADDITION AND SUBTRACTION OF MATRICES

have the same
Two matrices are said to be conformable for addition and subtraction if they
order.
If Aand B are two matrices of the same order, then their sum A+ Bis a matrix each
element of which is obtained by adding the corresponding elements of A and B.
+ b,
In general, if A=[alm xn and B= [b,)m xn; then A+ B= C= [cJm xnt Where c, Fa,-B is
Similarly, if Aand B are two matrices of the same order, then their difference A
correspond
a matrix whose elements are obtained by subtracting the elements of B from the
ing elements of A.
In general, if A=[a,]m xn and B= [b,)m xp then A-B=C= lcmxn Where c, F a, - b,
2 5- 1 1 -6 2]
For examnple, if A=3 0 4and B=-2 5
2+1 5 -6 -1+2| 3 -1 1
A+ B=
then 3-2 0+5 4+7 5 11
2-1 5 -(-6) 11 31
and A-B=
3-( 2) 0-5 -1-2-| -5 -3)
4-7

1.5. MULTIPLICATION OF A MATRIX BY A SCALAR

The product of amatrix A= [a] by ascalar kis denoted by kA and is obtained by multiplying
every element of A by k.
Thus, kA = [ka,)
,then kA = [kai kag kas
If A=
b b3 |kb, kb, kbs
In particular, -A =(-1) A:
A=
-b, -ba)
1.6. PROPERTIES OF MATRIX ADDITION

() Matrix addition is commutative i.e., A + B= B+A

(ii) Matrix addition is associative i.e., (A + B) + C= A + (B + C)
(iii) For any matrix A, there exists a null matrix O of the same order as A such that
A+00+A = A.
(iv) For any matrix A, there exists a matrix A of the same order as A such that
A +(-A) = ( A) + A =0.
 OF ENGINEERING
6 ATEXTBOOK
MATHEMATICS
ILLUSTRATIVE EXAMPLES

6 4
Example [Link] x, y, zand wif 2 w z+ w
t+4 6+x+ y
Sol. The given equation is 3x 3y
32 3w -1+2+ w 2w +3
Equating the corresponding elements on the two sides
32 =-1 +z+ w, 3w =2w +3
3x = x+ 4, 3y = 6 +x+y,
22 =- 1 + w, W=3
2x= 4, 2y = 6 + x,
x= 2, y=4, z=1, w =3.

2 5 -7
Example 2. Express 9 12 4 as the sum ofa lower triangular matrix and an
15 -13 6
upper triangular matrix with zero leading diagonal.
a 0 0]
Sol. Let L=|b c 0 be the lower triangular matrix and î =|0 0
r be the upper
d e f
00 0
triangular matrix with zero leading diagonal such that
2 5 -7 a 0 0
-9 12
0 p q a p q
4 =b c 0+0 0 r|
15 -13 6 d e f|0 0 0| d e f
Equating the corresponding elements on the two sides
2= a, 5=p,
-7=q,-9=b, 12 = c, 4=r, 15 = d, -13 =e. 6=f
2 0 01 fo 5 -7
L= 9 12 0 and U=0 0 4
15 -13 6 0 0 0
1.7. MATRIX MULTIPLICATION

Two matrices Aand B are said to be

A and B) if the conformable
number of columns in A(called thefor the product AB (in very order of
in B (called the post-factor). pre-factor) is egual to this
the number of rows
Thus. if the orders of A and B are m Xn
(i) AB is defined if and p x q
number of columns inA = number of respectively,
then
(ii) BAis defined if number of
columns in B=
rows in B, i.e.. if n =p
Let A=[a,mxn' and B= [b,J,xn be two number of rows in A, i.e., if q = m.
matrices conformable
for the product AB, then
ADis defined as the matrix C=
lcjmxp Where Cj a, b,t a,,b, t.. +
L.e., (i, jyth element of AB = sum of the a,,b,,=) aj
corresponding elements of jih column of B. products of the
elements k=1
of ith row of A with the
MATRICES 7

The rule for multiplication of two conformable matrices is called row-by-column

method.

b2
Consider A=a1 L2 ag3 and B= b1 ba
31 dg2 d33
Orders of A and B are 3 x 3 and 3 x 2 respectively. AB is defined and is of order 3 x 2.
C1 C12
AB =C=C21 C22
C31 C32
where C1= sum of products of elements of 18t row of A and 18t column of B

= [a,, ag a,3l b1 = a,,b, t aj2bg1 t a,gbg1

b1
C,9 = Sum of products of elements of 1st row of Aand 2nd column ofB

=a,,b, ta,pbgg t a,gbgp

Co1 = Sum of products of elements of 2nd row of A and 1st column oF B

be1 =a,,b, t a,gb t agsbg1

|bg1
Co9 Sum of products of elements of 2nd row of Aand 2nd column of B
by2
= [a,, ayg Cygl bg2 =azbgta,gbg, t agbgg
bg2.
Co1 = Sum of products of elements of 3rd row of A and 1st column of B

agl b1 =a,,b, t a,gbg t agsba

b1
C32 = sum of products of elements of 3rd row of A and 2nd column of B
[b2
b2 = a,jb t aggbgg t aggbg2

abnt aj2byn t ajsba1 anbi t apbga t a1sbs2

Thus, AB=a,b1 t aggbg1 t+ aggbai agnb12 t aggbg t agsbs2
1 -2 31 [1 0 21
Example [Link] = 2 3-1| and B= 0 1 2 Obtain the products AB and BA
-3 1 2 1 2 0
and show that AB # BA.
1 -2 31||1 0 2]
Sol. AB = 2 3 -10 1 2
3 1 2||1 2 0
 ATEXTBOOK OF ENGINEERING
8

1(1) 2(0) + 3(1)

1(0) 2(1) + 3(2)
1(2) 2(2) +3(0)7 4 4 -2
2(2) + 3(2) 1(0) = 1
MATHEMATICS
2(0)++3(1) - 1/2)
2(1)++3(0)
|3(1) 1(0) +1(1)
2(1) -3(0) 1(1) + 2(2) - 3(2) + 1(2) 4+2(0)-1 1
5 -410

1 0 21T 1 -2 31
BA=0 1 2 2 3 -1
1 2 0|-3 1 2
|1(1) + 0(2) + 2(-3) 1(-2)+ 0(3) + 2(1) 1(3) + 0(- 1)+ 2(2) [-5 0 71
0(1) + 1(2) + 2(-3) 0(-2) + 1(3) + 2(1) 03) + 1- 1)+ 2(2) =-4 5
1(1) + 2(2) +0(-3) 1(-2) +2(3) + 0(1) 1(3) + 2(- 1) + 0(2) 5 41
Orders of AB and BA are the same but their corresponding elements are not equal.
Hence AB BA.

1.8. PROPERTIES OF MATRIX MULTIPLICATION

) Matrix multiplication is not commutative in general i.e.,AB BA.
() Matrix multiplication is associative i.e., (AB)C =A(BC).
(iüi) Matrix multiplication is distributive with respect to matrix addition i.e.,
A(B+ C)= AB + AC.
(iv) If A and Iare square matrices of the same order, then AI = A= A.
(v) If Ais a square matrix of order n, then AxA = A2, A x A XA = A, A x A
times = Am are all square matrices of order n. xA ... m
Also, we define A° =I.
(vi) For any positive integer n, I" =I.

1 -2 31
Example 4. Evaluate A'- 3A +91, ifI is the unit matrix of order 3 and A= 3 -1
-3 1 2
1 -2 3| 1 -2
Sol. A2 = AxA= 2 3-1 2 3-1
31-12 -5 11
11 4 1
-3 1 2|-3
1 2 -7 11 6|
[-12 -5 11] 1 -2
A2 3A+ 9I, = 11 4 13 2
3 1 0 0]
-7 11 6
3 -1+90 10
-3 1 2
0 0 1
-12 -5 11 3 -6
11 4 9 9 0 0 -6 1 21
1 6
9-3+0 9 0
-7 11 -6-9 3
5 4 4
60 0 9 2 8 -3
1.9. MATHEMATICAL INDUCTION

Mathematical induction is a very useful device

the result to be proved involves n, where nis a for proving results for all positive integers. I!
induction consists of the following two steps: positive integer, then the proof by mathematical
Step 1. Verify the result for n = 1.
Step 2. Assume the result to be true for n =k
and then prove that it is true for
n=k+ 1,
MATRICES

Now the result is true for n = 1

Using step 2, the result is true forn=1+1 =2
The result is true for n=2+ 1=3
The result is true for n =3+ 1=4 and so on.
Hence, the result is true for all positive integers n.
Example 5. By mathematical induction, prove that if
[3 -4 then An I+ 2n -4n where n is any positive integer.
A=1 -1 1-2n
Sol. We prove the result by mathematical induction.
1+2n - 4n |
When n = 1, A" = n 1-2n
1+2.1 -4.1
A' = 1 1-2.1
’ The result is true when n = 1.
Let us assume that the result is true for any positive integer k.
1+2k-4k] ...(1)
L.e., let A= 1-2k
1+2k -4k 13 4
[Using (1)]
Now, Ak*l =A". A=1- 28 1 -1
3(1+ 2k) 4k - 4(1+ 2k) +4k|3+2k -4-4k
T3k +1- 2k -4k-1+2k 1+k -1- 2k
1+2(k +1) 4(k + 1) |
=k+1 1-2(* +1)
The result is true for n = k +1
Hence, by mathematical induction, the result is true for all positive integers n.
1.10. TRANSPOSE OFA MATRIX
Given a matrix A, then the matrix obtained from A by changing A'its rows into columns and
by or A'.
columns into rows is called the transpose of A and is denoted
1 2
02 0 -1
-1 3 5then
1
For example, if A= A=
7
5 7

Clearly (i) if the order of A is m xn then order of A' is n x m.

() (i, j)h element of A' = G, iyth element of A.
In symbols, if A= [a, ]m. xnt then A' = [b,,xm Where b, = a,:
1.11. PROPERTIES OF TRANSPOSE OF A MATRIX
If A' and B' denote the transposes of A and B respectively, then
() (Ay= A ie., the transpose of the transpose of a matrix is
the matrix itself.
10
ATEXTBOOK OF
ENGINEERING
(ii) (A + B)' = A' + B'ie, the transpose of the sum of two matrices
sum of their is
MATHEMATIC
equal tothe
S
transpOses.
(i) (AB)' = B'A' ie., the transpose of the product of two matrices
product of their transposes taken in the reverse order. is equal to the
1.12. SYMMETRICMATRIX
A square matrixA=la. is said to be
equal to the matrix itself.
symmetric if A'= Ai.e., if the transpose of the matry
Thus, for a symmetric matrix A = [a, a,
Fa,i

[a hg
For example, 6 |h b f are symmetric matrices.
g f c

1.13.
SKEW-SYMMETRIC MATRIX (OR ANTI-SYMMETRIC MATRIX)
A square matrix A = ka,} is said to be
matrix is skew-symmetric if A'=-A i.e., if the transpose of the
equal to the negative of the matrix.
Thus, for a
Putting j=i,skew-symmetric
a,=-a,; ’
matrix A= |a, a, =-ai
Thus, all diagonal elements of a2a; =0 or a,, =0 for all i.
2 -30 skew-symmetric matrix are zero.
For example, 2 0 1 h -g
-h 0 are skew-symmetric matrices.

1.14. EVERY SQUARE MATRIX CAN

3-1 0 g -f 6
A
SYMMETRIC MATRIX ANDAUNIQUELY BE EXPRESSED AS THE SUM oF
Let A be any square matrix. SKEW-SYMMETRIC
MATRIX
Consider B=}A+ A)and C=}A- A)
B+C=A+ A) + A- A) =A
Also
B=(A +Ay = A' + (AY] =}(A'+ A)
’ Bis a symmetric =B
matrix.

Cis a
C=}A-A)' =}[A- (AY] = [A- A]
=-}A- A) =-C
skew-symmetric matrix.
Hence every square mnatrix A can be
is a symmetric matrix and C = expressed as A=B+ C. where B=(ATA)
2 (A- A) is a
To prove that there skew -symmetric matrix.
1s only one way in
metric matrix and a which A
skew-symmetric matrix, suppose can be exnressed as the sum of a Sy
MATRICES 11

A=P+Q ...(1)
is another such representation in which P is a symmetric matrix, and Q is a skew-symmetric
matrix.
i.e., P'= Pand Q=Q
Then A'= (P+ QY=P'+ Q= P- Q ...(2)
From (1) and (2), A + A'= 2P and A A' = 2Q
P=A+ A) =BandQ= A- A) = Cso that P+Q=B+C
Hence the result.

1.15. ORTHOGONAL MATRIX

Asquare matrix Ais called an orthogonal matrix if AA' =A'A =I or A' =A.
Note. If Aand B are any two orthogonal matrices, AB will also be an orthogon al matrix.
1.16. NILPOTENT MATRIX

Asquare matrix A is said to be nilpotent if A' = 0 e.g.

A-8
It will be of index p ifp is the least positive integer such that AP =0.
1.17. IDEMPOTENT MATRIX

1 0
A square matrix Ais idempotent if A' =A eg. A=
A square matrix A is said to be idempotent of period n ifn is the least positive integer
such that An+1 = A.

1.18. INVOLUTARY MATRIX

A square matrix A is said to be involutary ifA2 = I where I is a unit matrix.

-5 -8 0
e.g. A= 3 5
1 2

Example 6. IfA and B are symmetric matrices, proue that AB - BAisa skew-symmetric
matri.
Sol. A and B are symmetric matrices.
A' = A and B' = B ...(1)
Now, (AB BAY = (ABy' -(BAY' [: (A - BY = A' - B]
= B'A'- A'B' [:: (AB)' = B'A]
= BA - AB (Using (1))
=-(AB- BA)
. (AB- BA)is a skew-symmetric matrix.
 12
symmetric and a.
Example 7. Expresss the followving matrix
asthe sum ofa skew-symmetric
matrix,
[-17 1|
2 3 4
5 05
-1 2 5]
[-1 7 11
A'= 73 0
Sol. Given matrix is A = 2 3 4 1 4 5
50 5
-1 9/2 31
B=
1 9/2 3 2
Now, (A + A) = 2 5
2 3

1
0 5/2 -2
C=,(A- A)=-5/2 0 2
2 -2 0

-1 9/2 31 0 5/2 -21

A=B+C=9/2 3 2+-5/2 0 2
3 2 5 2 -2

where B is symmetric and C is a skew-symmetric matrix.

1.19. ADJOINT OF A SQUARE MATRIX
The adjoint of a square matrix is the transpose of the matrix obtained by replacing each
element of Aby its co-factor in | A |. Adjoint of A is briefly written as adj. A.
bË
Thus, ifA =ay b2 C2 then adj. A = transpose of A B1 C1 [A1 Ag Ag
b3 C3 B2 BË B2 Bs
|Ag B3 Cg C, C2 C3.
where the capital letters denote the co-factors of
corresponding small letters in
|A|= b, C2
b3 C3
Note. If A is an n-rowed square matrix then
A (adj. A) = (adj. A) A = |A| I,:
1.19.1. Properties of Adjoint
()1HA and B are two non-singular square matrices of the same order then
adj. (AB) = (adj. B) (adj. A).
(2) If Ais a non singular matrix of order n then,
() |adj. A|=|Al-l (ii) adj. (adj . A) =|A n-2 A (ii) adj. AT = (adj. A)T.
1.20. INVERSE (or Reciprocal) OF A SQUARE MATRIX
Let A be an n-rowed square matrix. If there exists an
n-rowed square matrix B such that
AB= BA =I
MATRICES 13

then the matrixA is said to be invertible and B is called the inverse (or reciprocal) of A.
The inverse of a square matrix if it exists is unique. The necessary and sufficient condition for
a square matrix Ato possess inverse is that A is non-singular i.e., |A|#0.
Note 1. Inverse of Ais denoted by A-', thus B =A- and AA-= A-'A =I.
Note 2. Al = adj. A :|A|* 0. Note 3.|A-|=|A .
|A|
1.21. PROPERTIES OF INVERSE

(1) IfA is invertible then so is A- and (A-1y-l = A.

(2)If A and B are two non-singular square matrices of the same order then
(AB)-1 = B-1 A-1. (U.K.T.U. 2011)
(3) If Ais a non-singular square matrix then so is A' and (A l= (Ay.
3-3 4
Example 8. IfA=2 -3 4,() find A-1 (ii)show that A3 =A-l,
0 -1 1
3 -3 4
Sol. (i) |A|=2 -3 4 =1 A-l exists
0 -1 1
1 -1 01
adj. A=-2 3 -4
-2 3 -3|
adj. A 1 -1 0
A-1= -2 3 -4
|A| -2 3 -3
3 -3 4|3 -3 4] 3-4 4
(iü) A2 =A xA=2 -3 4| 2 -3 4= 0-1 0
0 -1 1||0 -1 1 -1 2 -3

3 -4 41 3 -4 4] 1 0 01
A= A2 x A2 = 0-1 0 -1 0 =0 1 0=I
-1 2 -3|-1 2 -3 0 0 1
Now, A- A = AS A =I A3 = A-1,

TEST YOUR KNOWLEDGE

1. IfA is a square matrix, prove that

(i) A + A' is symmetric. (iü) A A' is skew-symmetric.
2. Prove that
(i) if A, Bare symmetric, then so is A+B (ii) if A, Bare skew-symmetric, then so is A+B
(iiyif A is a square matrix, then AA' and A'A are both symmetric.
(iv) if Ais symmetric, then B AB is symmetric.
 TEXTBOOK OF ENGINEERING
14
A
MAATHEMATICS
[1 2 1]
3. If A= a 0 4 and adj. (adj. A) = A,
find a. (M.T.U. 2012
|1 1 1|
1 2 -1] [1 0 0] verify that (AB)'= BA'.
4. If A=3 0 2 and B=2 1 0.
4 5 0 01 3
are orthogonal.
6. Prove that the following matrices
[-8 4 1|
[-2 1 2] 1 4 -8
2 2 1 4 7 4
6. Express1 each
-2 of
2 the following matrices as the sum of a symmetric and a skew-symmetric

matrix:
4 2 - 3 a ab
1 3 -6 (ii) c bb
-5 0 -7 c a C

Answers
3. 3
4 2
-4 0 1
6. (2) olN 3 3 -3+ 0 -3
4 -3 -7 -i 3
la +o) (6+ c)] la-o) 6 -c)
( i ) a+c) b a +b) + - a) (b- a)
6+ c) a +b) C
3e- b) a-b) 0

1.22. ELEMENTARY TRANSFORMATIONS (or Operations)

Any one of the following operations on a matrix is called an elementary transformation
(or E-operation).
(i) Interchange of twO rows
or two
The interchange of ith and jih rows is columns.
The interchange of ith and j'h denoted by R::
columns isof)denoted
(ii) Multiplication of (each element a row orby column
C:* by a non-zero numberk.
The multiplication of i'h row by k
is
of ih column by kdenoted
The by R.(k).
multiplication
(iii) Addition of k is denoted by C,(k).
times the
elements of another row (or column), elements of a row (or column) to the Corresponding
k÷ 0.
The addition of k times the
jth row to the ith row is
The addition of k
times the jth column to the ith denoted by R,(R).
Ifa matrix Bis column is denoted by C,(). Bissaidto
t
be equivalent to A. Two obtained from a matrix A
equivalent matrices AbyandoneBorare more E-operations,
then

written as A ~ B.
MATRICES 15

1.23. ELEMENTARY MATRICES

The matrix obtained froma unit matrix Iby subjecting it to one of the E-operations is called
an elemnentary matrix.

[1 0 0]
e.g Let I=0 1 0
00 1
1 0 0]
Operating Rgg 0r C¡, 0n I, we get the elementary matrix 0 0 [Link] is denoted by E,.
01 0
1.24. THEOREMS ON THE EFFECT OF E-OPERATIONS ON MATRICES
(a) Any E-rowoperation on the product of two matrices is equivalnt to the same E-row
operation on the pre-factor.
If the E-row operation is denoted by R. then R(AB) = R(A). B.
(b) Any E-column operation on the product of two matrices is equivalent to the same
E-column operation on the post-factor.
Ifthe E-column operation is denoted by C. then C(AB) =A. C(B).
(c) Every E-row operation on a matrix is equivalent to pre-multiplication by the
corresponding E-matrix.
Thus, the effect of E-rowoperation R,, on A= E,.A
The effect of E-row operation R,(k) on A= E,). A
The effect of E-row operation R(k) on A = E(k) . A.
(d) Every E-column operation on a matrixis equivalent to post-multiplication by the
correspon ding E-matrix.
Thus, the effect of E-columnoperation C,on A=A.E,
The effect of E-column operation C(%) on A=A. (E0)}
The effect of E-column operation C,(k) on A=A. E{).
(e) The inverse of an elementary matrix is an elementary marix.

ILLUSTRATIVE EXAMPLES

I 3 31
Example 1. Transform 2 4 10 intoa unit matrix by using elementary transjornations.
3 8 4
(U.K.T.U. 2011)
1 3 31
Sol. We have |2 4 10
3 8 4|
 OperatingI )
01-2
0-1-5
Operating R,c9, R,1)
[1 0 97
01-2

Operating R,-)
1 0 9
01 -2
Operating R,-9), R(2) 1
1 0 0
01 0
0 0 1
-1 2 -2
Example 2. Reduce the matrix 2
1| to the upper triangular form by usNE
-1 -1
elementary transformations.
Sol. We have -1 2 -21
1 2 1
Operating R,(), R,-1) -1 -1

2 -2
4 -1
Operating R1) 0-3 2

-1 2-2
0 4 -1
Operating R, ) 0 1
1

-1 2 -2

1.25. INVERSE OF
MATRIX BY 00 5/4
The elenentary row
transformat
applied to the unit matrix, EOPERATIONS (Gauss-Jordan
giveaions which reduee square
Method)
the
inverse matrix aA, matrix Ato the unit
matrix, when
MATRICES 17

To compute the inverse of a matrix, we use the concept of equivalent matrices.

If we are to find out the inverse of a non-singular square matrix A, we first write A as
equivalent to I, a unit matrix of the same order.
A-I
Then we apply elementary row operations on them. The objective is to reduce Ato I. As
soon as this is achieved, the other matrix gives A.
I~A-1
This is an elegant way of determining the inverse or reciprocal of a matrix A.
Bxample 3. Employing elementary row transformations, find the inverse of the matrix
0 1 2|
A= 1 2 3 (A.K.T.U. 2018)
3 1 1

Sol. Let A-I

fO 1 21 [1 0 0]
1 2 3 -
l0 1 0|
3 11 0 0 1
Operating R2
[1 2 3] fo 1 01
0 1 2 0 0
3 1 1 0 0 1
Operating Rg(-3)
1 2 3 1 0
1 2 -1
0 -5-8 |0 -3
Operating R- 2), R(5)
[1 0 -1 -2 1 0]
01 2 -
1 0 0
0 0 2 5 -3 1|

Operating R,g), Rg3- 1)

[1 0 0]
|0 1 0 4 3 -1
00 2 5 -3

Operating R,4)
[1 0-0]
0 1 0-4 3-1
0 0 15/2 -3/2 1/2
 ATEXTBOOK OF
18
ENGINEERING
I-A1

Al=-4 1
MATHEWATICS
|5/2 -3/2 1/2
Example 4. Find theinverse of the matrix: Mby applying elementary
M=
021 3
1 1 -1 -2
12 1
transformations
-I 1 2 6
Sol. Let
M- I,
[o 2 1 3] 1 0 0 0
1 1-1 -2 010 0
1 2 0 1 00 1 0
-1 1 2 6 000 1
Operating Rg
11 -1 -2] o 1 00]
02 1 3 100 0
1 2 0 001 0
-1 1 2 000 1
Operating R,- ), R,()
11 -1 -2] fo 1 0 0]
0 2 1 3 |1 0 00
01 1 3
0-1 1 0
02 1 4 0 1 0 1|
OperatingR-). R)
1 2 0 0
0 2 1 31 1 010]
000
0 11 3 0
00 01-1 -1101 01
Operating R -)
f1 2 0 11 o 0
010 01 1 0
01 1 3 11-10
0-1 10
0 1

100
00 -2 -2 3 0]
11-10
-22 20
1
 19
MATRICES

Operating R,,(-1), R-3)

1 0 0 0-1 -3 3-1]
0 10 0 1 1 -1 0
|0 0 10 2 -5 2 -3
0 0 0 1 1 1 0 1
I~ M-1
-1 -3 3-1
M-1=
1 1 -1
Hence, 2 -5 2 -3
-1 1 0 1

TEST YOUR KNOWLEDGE

[1 2 3]
1. Reduce the matrix Ato triangular form, A =2 5 7
3 1 2

2. If X, Yare non-singular matrices and B= show that

where O is a null matrix.

[1 0 ][1 0 0][1 0 0]
3. Find the inverse of the matrix 0 1 0|o 1 s 0 p 0without first evaluating the product.
0 0 10 0 10 0 1
4. Find the inverse of the following matrices by using elementary row operations:
[2 0 - 1] 2 1 -1| 8 4 3]
(i)5 1 (ii) 0 2 1 (iin)2 1 1
0 1 3 5 2 -3 |1 2 1|
3-1 11 1 2 1] 1 0 -1
(iv) -15 6-5 (u) 3 2 3 (ti)3 4 5
5 -2 2 1 1 2) 0 -6 -7
[3 -3 41 f1 2 5] 1 2 -2
(vii) 2 -3 4
0-1 1|
(viii) | 2 3 1 is)-10-23
-1 1 1
(A.K.T.U. 2014, 2019)
1 3 3] [1 2 1] 1 -1 1)
()1 4 3 (xi) 3 1 2 (vii)4 1 0
1 3 4 0 1 2 8 1 1
1 2 31 2 3 4] 2 4 3 2
(xiii)2 4 5 (xir) 4 3 1|(A.K.T.U. 2021) (ve)
3 6 5 2
3 5 6| 2 5 2 -3
1 2 4| 4 5 l4 14
1 1 1
3 5 7
1 1 i -1 2i]
(rvi) (xviü) 2 0 2
7 11
1 -1 0 1
11 13
 ATEXTBOOK OF ENGINEERING

following non-MATHsEiMnAgulTICaSy
20
of the
row transformations, find the inverse
5. Employing elementary
matrices:
2
[1 -1 0 21 1 -1 2
1 2 3 1| 0 1 1 -1 1 3 2 -3
1 33 2 (üi) 1
() 2 4 3 3
1 1 11
(i)\2 12
|3 -2 1 6
1
2 -3 -1
2 1 -1
4

Answers
1 2 31 [1 0 - k]
1 . 01
0 0 -2
1
3.o0 1/p0 -s/p1
3 -1 1 8 -1 - 31 1 -2 -11
4. ( ) -15 6 -5 ()-5 1 2 3
1 -5 2
5 -2 2 10 -1 -4 -3 12
2 0 -1 -1/4 3/4 -1] 1/10 3/10
(iv)5 1 1/5)
(v) 3/4 - 1/4 (vi) 21/20 -7/20 - 2/5
0 1 3 |-1/4 - 1/4 1 -9/10 -3/10 1/5
1 -1 0 2 3 -131 3 2 6]
(vii) 2 3-4 (vii) -3 6 (ix) 1 1 2
2 3 3 21 5 -3 -1| 2 2 5
[7 -3 -3 3 -31
() 1 1 0
1 2 -1]
(xi) 6 -2 -1 (ii) 4 -7 4
-1 0 1 S-1 1 -5| 4 -9 5
1 -3 2] - 10 4 97
(iii) 3 3 -1 1
15 -4 -14
2 -1 0 5
-5 1 6|
23 29 -64/5 - 18/5
10 - 12 26/5 7/5 55125
(xU) 1 -4851O 45045
1 -2 6/5 2/5| (Ui) 48510 105875
2 2238 35035
-2 3/5 45045
1/5 -35035 154154

(zvii)1_-4013i -2
2i
4
01 2

1 -2 1 0
2 -1
5.
1 -2 2 -3 1 -1 2 5-7 1
() 0 1 -1 1 (ii) -5 -3 1 1
2 3 -2 3
2 3 -1 0 (ii)1 5 -1 5 -2
18-7 5 11 10
-3-1 0 1 | 1 -2 10 5
1.26. RANK OF A MATRIX

Let Abe any m X n matrix. It has square

these square sub-matrices
sub-matrices are called minors
A matrix A is said to be of rank r if
of A
of different orders. The determinants of

() it has at least one

non-zero minor of order r
(i) all the minors of order (r+ 1) or
Symbolically, rank of A = r is higher than r are
written as p(A) =r. zero.
MATRICES 21

From the definition of the rank of a matrix A, it follows that:

() If Ais a null matrix, then p(A) =0. every minor of A has zero value.]
(ii) If A is not a null matrix, then p(A) > 1.
(iüi) If A is a non-singular n Xn matrix, then p(A) = n.
|:: |A | 0 is the largest minor of A.]
If I, is thenxn unit matrix, then | I, | =1z0 ’ p) =n.
(iv) If Ais an m x n matrix, then p(A) s minimum of m and n.
(v) If all minors of order r are equal to zero, then p(A)<r.
1.27. METHODS OF FINDING RANK
To determine the rank of a matrix A, we adopt the following different methods:
(1)Start with the highest order minor (or minors) of A. Let their order be r. If any one
of them is non-zero, then p(A) =r.
If all of them are zero, start with minors of next lower order (r- 1) and so on tillyou
get a non-zero minor. The order of that minor is the rank of A.
This method usually involves a lot of computational work since we have to evaluate
several determinants.
(2)Normal form method. If Ais anm xn matrix and by a series of elementary(row or
column or both) operations, it can be put into one of the following forms (called normal or
canonical forms):
I, : 0] [I,
..1, : 01, [],where I, is the unit matrix of order r.
Since the rank ofa matrix is not changed as a result of elementary transformations, it
follows that
p(A) =r. [:: th order minor | I, | =l÷0]
This method is also called sweep out method or pivotal method.
Procedure to obtain normal form
() Interchange rows (or columns) to obtain a non-zero element in I row and I column of
given matrix.
(ii) Make this non-zero element as 1.
(ii) Obtain zeros in the remainder of I row and I column.
(iv) Repeat the above three steps starting with element in II row and II column.
(U) Continue the process down the main diagonal either until the end of diagonal is
reached or all the remaining elements of matrix are zero.
(3) Echelon form method. In this form of the matrix, each of the first relements of
the leading diagonal is non-zero and every element below this diagonalrth row is zero. Amatrix
is reduced in Echelon form as
() The first non-zero element in a row should be unity (if possible).
() All the non-zero rows, if any, precede the zero rows.
The rank of the matrix is equal to the number of non-zero diagonal elements or the
number of non zero rows (as the case may be) when it has been reduced to Echelon form.
In other words, a matrix A= [a,] is an Echelon matrix or is said to be in Echelon
form if the number of zeros preceding the first non-zero entry of a row increases row by row
until only zero rows remain.
 ATEXTBOOK OF ENGINEERING
22

entry ln a rowreduced
of row Echelon form (row canonical formn) of the matrix the first
is unity and are the only non-zero entry in their respective columns.
MATHEMATIC
Further, any matrix can be transformed to a column Echelon
form
that a matrix is in CEF ir
by(CEF) non-Z6
of elementary column
transformations.
() allzero columns are on the right
We say
sequeno
first Gf we go down from the top) non-zero entry in each column is 1
(i) the
appears below that of c.
(i)) ifj>i, then the leading one of the column c,
[1 0 0] fo 0]
e.g.,
0 1 0
1
|3 2 0 |2 1
A matrix is said to be in a column reduced Echelon form ifit is in CEF and
any row containing the leading one of a column consists of all zeros except this leading one. In
above examples,first and third matrices are in column reduced echelon form.
additionally.
1.28. FOR AN mxnMATRIX A, TO FIND SQUARE MATRICES P AND Q OF ORDEDe
m AND n RESPECTIVELY, SUCH THAT PAQ IS IN THE NORMAL FORM
Method. Write A = IAI.
Reduce the matrix on LHSto normal form by affecting elementary row and/or column
transformations.
Every elementary row transformation on A must be accompanied by the same transfor
mation on the pre-factor on RHS.
Every elementary column transformation on A must be
formation on the post-factor on RHS. byaccompanied the same trans

1.29. RANK-NULLITY THEOREM

(A.K.T.U. 2021)
If A is an m x n order matrix
over some field then
rank (A) + nul
This applies to linear maps also. Let V(A) n
=
vector space over some field. Let T:V ’ W be a finite dimensional vector space ana "
be a linear map, then
rank (T) + nul (T) = dimn (V)
where the rank of Tis the
of the kernel of T. dimension of the image of T and nullity of T is the dimension

ILLUSTRATIVE EXAMPLES
Example 1. Find the rank of the matrix
1 2 3
(ii) |1 4 2

2 3 4 -11
|2 6 5
2 3 -1 -1|
(iüi) 5 2 0 -1|
4 5 12 -1 (iv) 1 -1-2 -4

6 3 3-2
0 -7
MATRICES 23

1 2 3 4]
Sol. (i) Here,
-2 0 5 7
Operating R, (2)
1 2 3 4]
4 11 15|
which is Echelon form.
p(A) = number of non-zero rows = 2
12 3
(ii) Here, A= 1 4 2
2 6 5
|A |= 0 . pA) = 3
1 2
A 2 x 2 order minor is
1 4 =2 0
p(A) = 2 I By definition
2 3 4 -1
(iüi) Here, A = 5 2 0 -1
-4 5 12 -1
Operating R (- 2)
2 3 4
1 -4 -8
-4 5 12 -1|
2020perating R2
1 4 -8 1
2 3 4-1
-4 5 12 -1

Operating R - 2). Ra1 (4)

1 -4 -8 1|
~0 11 20 -3
|0 -11 -20 3
Operating R (1)
[1 -4 -8
0 11 20 -3
|0 0 0
which is Echelon form.
P(A) = 2
2 3 -1-1
(iv) Here A= 1 -1 -2 -4 is a 4 x 4 matrix.
1 3 -2 .. p(A)s4.
6 3 0 -7
24 ATEXTBOOK OF
ENGINEERING
Operating R2
[1 -1 -2 -41
2 3 -1 -1
MATHEMIC
3 1 3-2
6 3 0 -7
Operating Ca(1), C,2), C4)
[1 0 0 0]
2 5 3 7
3 4 9 10
69 12 17|
Operating R,-2),R,-3), R,-6)
[1 0 0 01
|0 5 3 7
0 4 9 10
0 9 12 17|
Operating Rgg 1), R8-2)
1 0 0 0]
|0 1 -6 -3
0 4 9 10
0 1 -6 -3|
Operating Cs6), C,3)
1 0 0 01
01 0 0
0 4 33 22
01 0 0
Operating Rpl 4), R 1)
1 0 0 01
0 1
00 33 22
[0 0 0
Operating C, 1

[1 0 0 01
010 0
00 1 22
0 0 0 0
Operating Cag-22)
10 0 :
01
01 0 :0
00 1: 0 O3x1
which is normal form. 0 00 o Oxs O1x1
p(A) = 3.
 25
MATRICES
matrix A to
Example 2. Use elementary transformations to reduce the following
triangular form and hence find the rank of A.
[2 3 -1 -11
1 -1 -2 -4
A = 3 1 3-2
6 3 0 -7
Sol. We have,
2 3 -1 -1
1 -1 -2 -4
A =
1 3 -2
6 3 0 -7
Operating R2
1 -1 -2 -41
2 3 - 1 -1
1 3 -2
6 3 0 -7
Operating Rg, (- 2), Rj(-3), R, (-6)
1 -1 -2 -4
0 5 3 7
0 4 9 10
9 12 17
Operating Ryg- 1)
[1 -1 -2 -41
0 1 -6 -3
0 4 9 10
0 9 12 17
Operating Rg- 4), Rg- 9).
1 -1 -2 -4
0 1 -6 -3
0 0 33 22
0 0 66 44|
Operating Ryg- 2)
[1 -1 -2 -41
0 1 -6 -3
0 0 33 22
0 0 0
which is upper triangular form.
p(A) = 3.
Example 3. Find the rank of matrix
2 3-2 4
A 3-212
3 2 3 4
-2 4 0 5