Matrices and Determinants: Animation 3.1: Addition of Matrix Source & Credit: Elearn - Punjab

version: 1.
CHAPTER
3 Matrices and
Determinants
Animation 3.1: Addition of matrix

Source & Credit: elearn.punjab
1. Quadratic Equations eLearn.Punjab 1. Quadratic Equations eLearn.Punjab
3. Matrices and Determinants eLearn.Punjab 3. Matrices and Determinants eLearn.Punjab
is called an m by n matrix (written as m % n matrix).

3.1 Introduction m % n is called the order of the matrix in (iii). We usually use capital letters such as A, B,
C, X, Y, etc., to represent the matrices and small letters such as a, b, c,... I, m, n,...,a11 , a12 , a13
While solving linear systems of equations, a new notation was introduced to reduce
,...., etc., to indicate the entries of the matrices.
the amount of writing. For this new notation the word matrix was irst used by the English
Let the matrix in (iii) be denoted by A. The ith row and the jth column of A are indicated
mathematician James Sylvester (1814 - 1897). Arthur Cayley (1821 - 1895) developed the
in the following tabular representation of A.
theory of matrices and used them in the linear transformations. Now-a-days, matrices are
used in high speed computers and also in
jth column
other various disciplines.
F
The concept of determinants was used by Chinese and Japanese but the Japanese  a11 a12  a1n 
a  a2 n 
a13  a1 j
 21 
mathematician Seki Kowa (1642 - 1708) and the German Mathematician Gottfried Wilhelm a22 a23  a2 j
Leibniz (1646 - 1716) are credited for the invention of determinants. G. Cramer (1704 - 1752)  a31 a32  a3n 
 
a33  a3 j
   
applied the determinants successfully for solving the systems of linear equations.   
 ai1 ai 2  ain 
A= (iv)
A rectangular array o f numbers enclosed by a pair o f brackets such as:
 
ith row D ai 3  aij
      
a  amn 
2 3 0   m1 am 2 am 3  amj
1 -1 4 
 2 -1 3   
 -5 4 7  3 2 6 
 
(i) or (ii)
 
 4 1 -1
The elements of the ith row o f A are ai1 ai2 ai3 ...... aij ...... ain
while the elements of the jth column of A are a1j a2j a3j .... aij ...... amj .
We note that aij is the element of the ith row and jth column of A. The double subscripts are
is called a matrix. The horizontal lines of numbers are called rows and the vertical lines useful to name the elements of the matrices. For example, the element 7 is at a23 position in
of numbers are called columns. The numbers used in rows or columns are said to be the
 2 -1 3 
the matrix  
entries or elements of the matrix.
The matrix in (i) has two rows and three columns while the matrix in (ii) has 4 rows and  -5 4 7 
three columns. Note that the number of elements of the matrix in (ii) is 4 % 3 = 12. Now we
A = [aij]m%n or A = [aij], for i = 1, 2, 3,...., m; j = 1, 2, 3,...., n, where aij is the element of the ith
give a general deinition of a matrix.
row and jth column of A.
Generally, a bracketed rectangular array of m%n elements
aij (i = 1, 2, 3, ...., m; j = 1, 2, 3, ...., n), arranged in m rows and n columns such as:
 a11 a12  a1n  Note: aij is also known as the (i, j)th element or entry of A.
a  a2 n 
a13
 21 
a22 a23
   
(iii)
 
 
 am1 am 2  amn 
The elements (entries) of matrices need not always be numbers but in the study of
am 3
matrices, we shall take the elements o f the matrices from _ or from C.
version: 1.1 version: 1.1
2 3
Note: The matrix A is called real if all of its elements are real.
 .a..11.... a12 a13 ....a. 14 
 a .....a... a........ a 
matrix A. For example,  21 22................ 23 24  , in the matrix the entries of the principal diagonal
 a31 a...32 a..33... a34 
Row Matrix or Row vector: A matrix, which has only one row, i.e., a 1 % n matrix of the . .
 ........ .... 
..
form [ai1 ai2 ai3 ...... ain] is said to be a row matrix or a row vector.
 ...41 . 
....
a . a42 a 43 a44
Column Matrix or Column Vector: A matrix which has only one column i.e., an m % 1 are a11, a22, a33, a44 and the entries of the secondary diagonal are a14, a23, a32, a41.
 a1 j 
a 
The principal diagonal of a square matrix is also called the leading diagonal or main
 2j diagonal of the matrix.
 a3 j  is said to be a column matrix or a column vector.
 
matrix of the form
  
 amj 
   
For example [1 -1 3 4] is a row matrix having 4 columns and  
Diagonal Matrix: Let A = [aij] be a square matrix of order n.
 is a column matrix having If aij = 0 for all i ≠ j and at least one aij ≠ 0 for i = j, that is, some elements of the
 
3 rows. principal diagonal of A may be zero but not all, then the matrix A is called a diagonal matrix.
The matrices
Rectangular Matrix : If m ≠ n, then the matrix is called a rectangular matrix of order m %
0 0
1 0 0   0 
n, that is, the matrix in which the number of rows is not equal to the number of columns, is 0 0
said to be a rectangular matrix. [ ], 
0 2 0  and 0

1 0
0 0
7 are diagonal matrices.
0 0 5   
0 2
0 0 0 4
 2 -3 0 
1 2 4 
 2 3 1  
For example;   and  3 -1 5  are rectangular matrices of orders 2 % 3 and 4 % 3
 -  Scalar Matrix: Let A = [aij] be a square matrix of order n.
 
1 0 4
0 1 2 If aij = 0 for all i ≠ j and aij = k (some non-zero scalar) for all i = j, then the matrix A is
respectively.
called a scalar matrix of order n. For example;
Square Matrix : If m = n, then the matrix of order m % n is said to be a square matrix of order
3 0
a 0 0 0 
n or m. i.e., the matrix which has the same number of rows and columns is called a square 0 0
7 0  0 
0  and 
0
0 7  
3 0
0 0
matrix. For example;
 ,
a are scalar matrices of order 2, 3 and 4 respectively.
1 1 2   0 a   
0 3
 2 5 0 3
[ 0 ],  -1 6  and  2 -1 8  are square matrices of orders 1, 2 and 3 respectively.
0
0 0
 
 3 5 4 
Let A = [aij] be a square matrix o f order n, then the entries a11, a22, a33, ...., ann form the Unit Matrix or Identity Matrix : Let A = [aij] be a square matrix of order n. If aij = 0 for all
principal diagonal for the matrix A and the entries a1n, a2 n-1, a3 n-2, ...., an-1 2 , an1 form the i ≠ j and aij = 1 for all i = j, then the matrix A is called a unit matrix or identity matrix of order
secondary diagonal for the n. We denote such matrix by In and it is of the form:
4 5
1 0  0 Transpose of a Matrix:
0 0  0 
0
 1
I n = 0 1  0
 
0 If A is a matrix of order m % n then an n % m matrix obtained by interchanging the rows
    and columns of A, is called the transpose of A. It is denoted by At. If [aij]m%n then the transpose
0 0 0  1  of A is deined as:
At = [a’ij]n%m where a’ij= aji .. for i = 1, 2, 3, ...., n and j = 1, 2, 3, ..... , m
1 0 0 
The identity matrix of order 3 is denoted by I3, that is, I 3 = 0 1 0   b11 b12 b14 
0 0 1  = b21 b22 b24  , then
b13
For example, if B = [bij]3%4 b23
b31 b32 b33 b34 
Null Matrix or Zero Matrix : A square or rectangular matrix whose each element is zero, is Bt = [b’ij]4%3 where b’ij = bji for i = 1, 2, 3, 4 and j = 1, 2, 3 i.e.,
called a null or zero matrix. An m % n matrix with all its elements equal to zero, is denoted
 b '11 b '12 b '13   b11 b21 b31 

by Om%n. Null matrices may be of any order. Here are some examples:
b ' b '23  b b 
0  0 0 0 0  =
=B  21  12 22 b32 
b '22
b '31 b '32 b '33  b13 b23 b33 
t
0 0 0  0 0    
[ 0] , [ 0 0 0] ,   ,  , 0  , 0 0 0 0  
b '41 b '42 b '43 
 
b14 b24 b34 

0 0 0  0 0  0  0 0 0 0 
   
Note that the 2nd row of B has the same entries respectively as the 2nd column of Bt and
O may be used to denote null matrix of any order if there is no confusion. the 3rd row of Bt has the same entries respectively as the 3rd column of B etc.
Equal Matrices: Two matrices of the same order are said to be equal if their corresponding Example 1:
entries are equal. For example, A = [aij]m%n and
1 0 -1 2   2 -1 3 1 
A =  3 1 2 5  = 1 3 -1 4  , then show that
B = [bij]m%n are equal, i.e., A = B iff aij = bij for i = 1 , 2 , 3 , ...., m, j = 1, 2, 3, ..... , n. In other words,
 
If and B
0 -2 1 6   3 1 2 -1
A and B represent the same matrix.
3.1.1 Addition of Matrices

(A + B)t = At + Bt
Two matrices are conformable for addition if they are of the same order.
The sum A + B of two m % n matrices A = [aij] and B = [bij] is the m % n matrix C = [cij] formed Solution :
by adding the corresponding entries of A and B together. In symbols, we write as C = A + B , 1 0 -1 2   2 -1 3 1  1 + 2 0 + (-1) -1 + 3 2 +1 
that is: [cij] = [aij + bij] A + B=  3 1 2 5+
 1 3 - 1 4= 
 
+3 1 +1 3 +2 -( 1) +5 4 
 
 0 -2 1 6   3 1 2 -1  0 + 3 -2 + 1 1+ 2 6 + (-1) 
 3 -1 2 3 
=  4 4 1 9 
where cij = aij + bij for i = 1 , 2 , 3 , ...., m and j = 1, 2, 3, ..... , n.
Note that aij + bij is the (i, j)th element of A + B.
 3 -1 3 5
6 7
3 4 3
 -1 A - B = A + ( -B )
4 -1
and (A + B) =  2
t
1 3
(1) = [aij] + [-bij] = [aij - bij] for i = 1, 2, 3, ..., m; j = 1, 2, 3, ..., n
  Thus the matrix A - B is formed by subtracting each entry of B from the
3 9 5
corresponding entry of A.
Taking transpose of A and B, we have
1 0 2 1 3
3.1.4 Multiplication of two Matrices
0 2  -1 3 1 
3
=
=A   and B t   , so
1 Two matrices A and B are said to be conformable for the product AB if the number
 -1 1  3 -1 2 
t
   
2 of columns of A is equal to the number of rows of B.
2 5 6  1 4 -1 Let A = [aij] be a 2%3 matrix and B = [bij] be a 3%2 matrix. Then the product AB is deined
1 3 0 2 1 3 3 4 3
to be the 2%2 matrix C whose element cij is the sum of products of the corresponding
0 1 -2   -1 3 1   -1 4 -1
elements of the ith row of A with elements of jth column of B. The element c21 of C is shown
A +B 
= + =   
 -1 2 1   3 -1 2   2 1 3 
t t
(2) in the igure (A), that is
     
2 5 6   1 4 -1  3 9 5 
From (1) and (2), we have (A + B)t = At + Bt
3.1.2 Scalar Multiplication
If A = [aij] is m % n matrix and k is a scalar, then the product of k and A, denoted by kA, is
the matrix formed by multiplying each entry of A by k, that is,
kA = [kaij]
Obviously, order of kA is m % n. c21 = a21b11 + a22b21 + a23b31. Thus
 b11 b12 
Note. If n is a positive integer, then A + A + A + .... to n times = nA.
==AB 
 a11 a12 a13     a11b11 + a12b21 + a13b31 a11b12 + a12b22 + a13b32 
  21 22   
 a21 a22 a23  b b   a21b11 + a22b21 + a23b31 a21b12 + a22b22 + a23b32 
b b
If A = [aij] U Mm%n (the set of all m % n matrices over the real ield _ then kaij U _, for all i
and j, which shows that kA U Mm%n . It follows that the set Mm%n possesses the closure property  31 32 
 b11 b12 
   a11 a12 a13 
with respect to scalar multiplication. If A, B U M and r,s are scalars, then we can prove that
Similarly BA = b21 b22  
a21 a22 a23 
r(sA) = (rs)A, (r + s)A = rA + sA, r(A + B) = rA + rB.
b31 b32  
 b11a11 + b12 a21 b11a12 + b12 a22 b11a13 + b12 a23 

3.1.3 Subtraction of Matrices
=+b a b a +b a b22 a22 +b21a13 b22 a23 
 
b31a11 + b32 a21 b31a12 + b32 a22 b31a13 + b32 a23 
If A = [aij] and B = [bij] are matrices of order m % n, then we deine subtraction of B 21 11 22 21 21 12
from A as:
8 9
AB and BA are deined and their orders are 2%2 and 3%3 respectively.
3.2 Determinant of a 2 % 2 matrix
Note 1. Both products AB and BA are deined but AB ≠ BA
2. If the product AB is deined, then the order of the product We can associate with every square matrix A over _ or C, a number |A|, known as
can be illustrated as given below: the determinant of the matrix A.
Order of A The determinant of a matrix is denoted by enclosing its square array between vertical
bars instead of brackets. The number of elements in any row or column is called the order
Order of B of determinant. For example,
a b
if A =  
a b
c d
Order of AB , then the determinant of A is . Its value is deined to be the real number
c d
 2 -1 0   2 -2 3 ad - bc, that is,

Example=    1 -4 6  , then compute A2B.
2: If A 1 2 -3 and B =-
   
1 2 -2   0 -5 5  = = ad - bc
a b
A
c d
Solution :
 2 -1 0   2 -1 0   2 -1 1 4 
A= AA 1- 2
= 3 1- 2 3 == if A   and B  2 8  , then
      
2 For example,
1 2 -2  1 2 -2 
4 3
2 -1
 4 - 1 + 0 -2 - 2 + 0 0 + 3 + 0   3 -4 3  A= = (2)(3) - (-1)(4) = 6 + 4 = 10
=  2 + 2 - 3 -1 + 4 - 6 0 - 6 + 6  = 1 -3 0 
   
4 3
 2 + 2 - 2 -1 + 4 - 4 0 - 6 + 4   2 -1 -2 
B= = (1)(8) - (4)(2) = 8 - 8 = 0
1 4
and
 3 -4 3   2 -2 3
2 8
∴ A2 B= 1 -3 0   -1 -4 6 
  
Hence the determinant of a matrix is the diference of the products of the entries in the
 2 -1 -2   0 -5 5  two diagonals.
6 + 4 + 0 -6 + 16 - 15 9 - 24 + 15 10 -5 0 
=  2 + 3 + 0 -2 + 12 + 0 3 - 18 + 0  =  5 10 -15 = ad - bc
a b
   
..
........
 4 + 1 + 0 -4 + 4 + 10 6 - 6 - 10   5 10 -10 
c d
..
-bc ad
Note: Powers of square matrices are deined as:
A2 = A % A, A3 = A % A % A, Note: The determinant of a 1%1 matrix [a11] is deined as |a11|= a11
An = A % A % A % .... to n factors.
10 11
3.2.1 Singular and Non-Singular Matrices

 ap + br ap + bs  1 0 
or  = 
 cp + dr cq + ds  0 1 
A square matrix A is singular if |A| = 0, otherwise it is a non singular matrix. In the above
1 4   2 -1
example, |B| = 0 ⇒ 2 8 ⇒  4 3  is a
  = ap += +
is a singular matrix and |A| = 10 ≠ 0
  ⇒
A =
br 1....(i) ap bs 0....(ii)
=cp += +
dr 0....(iii) cq ds 1....(iv)
non-singular matrix.
-c
From (iii), r = p
3.2.2 Adjoint of a 2 % 2 Matrix d
a b 
Putting the value of r in (i), we have
The adjoint of the matrix A =   is denoted by adj A and is deined as: adj  -c   ad - bc 
c d  ap + b  p  =1 ⇒   p =1 ⇒ p =
d
 d   d  ad - bc
 d -b 
A=  
 -c a  -c -c
r= p= = -
d c
d ad - bc ad - bc
and .
3.2.3 Inverse of a 2 % 2 Matrix d
Similarly, solving (ii) and (iv), we get

Let A be a non-singualr square matrix of order 2. If there exists a matrix B such that
-b
= =
a
1 0  ad - bc ad - bc
q , s
AB = BA = I2 where I2 =   , then B is called the
0 1 
 p q
Substituting these values in   , we have
r s
multiplicative inverse of A and is usually denoted by A , that is, B = A
-1 -1
Thus AA-1 = A-1A = I2

 d -b 
 ad - bc ad - bc  1  d -b 
=
=A-1  
Example 3: For a non-singular matrix A, prove that A = adjA  -c a  ad - bc  -c a 
1
 ad - bc ad - bc 
a b   p q
A
Solution : If A =   and A =  r s  , Then:

c d   
-1
Thus A-1 =
1
AA-1 = I2, that is, AdjA
A
a b   p q 1 0 
c d   r s  = 0 1 
     
12 13
3.3 Solution of simultaneous linear equations by using matrices

 5 3
Example 4: Find A-1 if A =   and verify that AA = A A
 
-1 -1
1 1
Let the system of linear equations be

Solution : A = =5-3=2 a11 x1 + a12 x2 = b1 
5 3

a21 x1 + a22 x2 = b2 
1 1 (i)
where a11, a12, a21, a22,+b1 and

a   x  b 
Since |A|≠ 0, we can ind A-1.
+ =  =  1=are realornumbers.
a
      
b
x1+written
AX = B (ii)
a x =
A-1 = + =  22 2  2  
1 The system (i) cana21be bin the matrix form as:
   
AdjA
 a11 a12   x1   b=  =
          =  1  or  AX  =B  
A
a    =     
 21 a22   x2 21 b222  2    2  -   
(ii)
1 -3 
a a x b
1  1 -3  2 2 a11 a12A ==  a -x1a=
where   b x    b 
⇒ =
A-1 = 
2  -1 5   -1
 where A =   = 
 
, X =   B X  1= B   
5
 a21 a22    x2   b2    -
 
 2 2  
   -    
 x  1  a  =-a  b    
=
 1 -3  If |A| ≠ 0, then A-1 exists
 


= - 
 -(ii) gives
    =  
- 
 by A-1)- 
or so
 5 3  2 2  - 
A. A-1 =     -  
(By pre-multiplying (ii)
 
A-1(AX) = A-1B
1 1  -1 5     isassociative)
Now
  
⇒ X = A-1 B =  (a =A-1A =  -I2=)  = 
-1 -1
or (A A)X = A B (Matrix multiplication
 2 2  - 
 -   -  
 5 3 -15 15  ⇒ X=A B   
= =  
2 - 2 + 
 
-1
1 0
==   -a12   b1 
-3 5  0 1  x  1  a
2 2
1 - 1 or  1  =  22  
(i)
+
 2 2 2 2   x2  A  -a21 a11  b2 
== = =
 1 -3   b1a22 - a12b2 
 2   5 3  
and A-1. A =  2  1  a22b1 - a12b2   
 ==  
 -1 5  1 1
A
A  -a21b1 - a11b2   a11b2 - b1a21 
 2 2   
 
5 3 3 3
A
2 2 - - 
1 0
==  
1 
b1 a12 a11 b1
 -5 + 5 -3 + 5  0
2 2
=
=
(ii) b2 a22 b21 b2
 2 2 2 2 
Thus and x2
A A
From (i) and (ii), we have It follows from the above discussion that the system of linear equations such as (i) has
AA-1 = A-1A a unique solution if |A| ≠ 0.
14 15
Example 5: Solve the following systems of linear equations.

x1 + 2 x2 = 4 
= = 3 x1 - x2 1 + x1 2 x2 4   is
  2 x1 + 4 x2 =12 
(ii) The matrix form of the system
= = x1 + x2 3 + 2 x1 4 x2 12 
i) ii)
 3 -1  x   1  1 2   x1   4 
    ==   == - = 3+x1 - x2 1 +       2  4  x  = 12 
   
  form of the system   is  
= = = x1 + x2 = 3 + x1 + x2 3 +      =   2   

i)
     1 2
Solution : (i) The matrix
 - 
=  3 -1  x13 - 11  x   1 
-  = = 1 2 A=
1 1   x  3     
 = = 4 - 4 = 0, so A-1 does not exist.
    2  
and A
-   3 -1  - 
2 4
  A =3 -1  =  x1   1
= = = A   -  , X -  and  B  
Multiplying the irst equation o f the above system by 2, we have
 1 1   x2   3
or AX = B . . . . (i) where
     
2x1 + 4x2 = 8 but 2x1 + 4x2 = 12

= -
= -  3 - -1   - 

which is impossible. Thus the system has no solution.
 A = - =3 + 1 = 4
-1 1     
   1- 1=  -     
Exercise 3.1
 =
  and  adj= A=   
= ,therefore,   -   
   - 1 3  -  -  -   2 3 1 7 
 =  =
= A   6 4  , then show that
  -      
   1 5  
1. If and B
    1  1 
  1  1  1  4  4   4A - 3A = A 3B - 3A = 3(B - A)
= =  =   
   4  -1 =3  1  3
i) ii)

-1
 =     -  -   -     
A
  i 0 
-      4  4  A= 
  1 -i 
, show that A4 = I2.
 (I)
2. If
 X =A-1B , that  is,
     
 1 1   = 
becomes
  = 
    
3. Find x and y if
 1
= 4- 4   -  
x   
1 3 3   x + 3 1   2 1 x + 3 1  y 1
x 1
 2  -   =
=
i)      -3 3 y - 4   -3 2 x 
 4 4   -3 3 y - 4   -3 2     
ii)
 1 3
 4 + 4  1 
= =     -1 2 3  0 3 2 
=
= A   and B 1 -1 2  , ind the following matrices;
- 1 + 9  2  1 0 2  
4. If
 4 4 
4A - 3B = A A + 3(B - A)
⇒ x1 = 1 and x2 = 2
i) ii)
2 0 x  1 x y   4 -2 3
Find x and y If   + 0 2 -1 =
 
1 y 3    1 6 1
5. 2
16 17
If A = [aij]3%3, ind the following matrices;

i) λ ( µ A ) = ( λµ ) A ( λ + µ ) A =λ A + µ A λ A - A = ( λ - 1) A
6.
3.4 Field
ii) ii)
If A = [aij]2%3 and B = [bij]2%3, show that λ ( A + B) = λ A + λ B.

A set F is called a ield if the operations of addition ‘+ ’ and multiplication ‘. ’ on F satisfy
7.
the following properties written in tabular form as:
1 2 0 0 
=
=
Addition Multiplication
8. If A   0 0  , ind the values of a and b.
a b   
and A2 Closure
i) For any a,b U F, For any a,b U F,
 1 -1 1 0 
a+bUF a.b U F
=
=
9. If A   0 1  , ind the values of a and b.
Commutativity
a b   
and A2
ii) For any a,b U F, For any a,b U F,
a+b=b+a a.b = b.a
1 -1 2  2 3 0 
=
=
10. If A   and B 1 2 -1 , then show that (A + B) = A + B .
Associativity
   
t t t
0 3 1 iii) For any a,b,c U F, For any a,b,c U F,
(a + b) + c = a + (b + c) (a.b).c = a.(b.c)
1 1 3
Existence of Identity
11. Find A3 if A =  5 2 6 
iv) For any For any
 
 -2 -1 -3
a U F ,\ 0 U F such that a U F ,\ 1 U F such that
a+0=0+a=a a.1 = 1.a = a
12. Find the matrix X if;. Existence of Inverses
v) For any a U F, a ≠ 0
 5 2   -1 5  5 2 2 1 
v) For any
a U F ,\ -a U F such that
 -2 1  X =  5 10 
∃ U F such that
i) X   
1
 -2 1  12 3    
ii)
a=
.( ) (=
a
1 1
a + (-a ) = (-a ) + a = 0 ).a 1
a a
Distributivity
13. Find the matrix A if,
vi) For any a,b,c U F, D1 : a(b + c) = ab + ac
D2 : (b + c)a = ba + ca
5 -1  3 -7 
 2 -1 0 -3 8 
i) 0 0  A =  0 0 
All the above mentioned properties hold for Q, _, and C.
     -1 2  A =  3 3 -7 
   
ii)
 3 1  7 2  3.5 Properties of Matrix Addition , Scalar Multiplication and
 r cos φ 0 - sin φ   cos φ sin φ 
Matrix Multiplication.
14. Show that  0 0  0 0  = rI 3 .
0
  
r 1
 r sin φ 0 cos φ   -r sin φ 0 r cos φ 
If A, B and C are n%n matrices and c and d are scalers, the following properties
are true:
18 19
1. Commutative property w.r.t. addition: A + B = B + A

Note: w.r.t. is used for “with respect to”. 1 × 2 + (-1) × 1 + 0 × 3 1 × 0 + (-1) × 4 + 0 × 0 1 × 1 + (-1) × 2 + 0 × 6 
Associative property w.r.t. addition: (A + B) + C - A + (B + C) =  2 × 2 + 3 × 1 + (-1) × 3 2 × 0 + 3 × 4 + (-1) × 0 2 × 1 + 3 × 2 + (-1) × 6 
 
1 × 2 + (-2) × 1 + 3 × 3 1 × 0 + (-2) × 4 + 3 × 0 1 × 1 + (-2) × 2 + 3 × 6 
2.
3. Associative property of scalar multiplication: (cd)A = c(dA)
Existance of additive identity: A + O = O + A - A (O is null matrix) 1 -4 -1
=  4 12 2 
4.
 
5. Existance of multiplicative identity: IA = AI = A (I is unit/identity (2)
matrix) 9 -8 15 
6. Distributive property w.r.t scalar multiplication:
(a) c(A + B) = cA + cB (b) (c + d)A = cA + dA Thus from (1) and (2), AB ≠ BA
7. Associative property w.r.t. multiplication: A(BC) = (AB)C
8. Left distributive property: A(B + C) = AB + AC Note: Matrix multiplication is not commutative in general
9. Right distributive property: (A + B)C = AC + BC
10. c(AB) = (cA)B = A(cB)  2 -1 3 0 
= 2: If - 1 0 4 2  ,then ind AAt and (At).
 
Example
2 0 1 1 -1 0  -3 5 2 -1
= 1: Find AB and BA if A -1 4 2  and B =  2 3 1
   
Example
 3 0 6  1 -2 3  Solution : Taking transpose of A, we have
 2 1 -3
 2 0 1  1 -1 0  -1 0 5 
=   1 A =  , so
Solution : AB -1 4 2   2 3  3 4 2
t
 3 0 6  1 -2 3   
 0 -2 -1
 2 × 1 + 0 × 2 + 1 × 1 2 × (-1) + 0 × 3 + 1 × (-2) 2 × 0 + 0 × ( -1) + 1 × 3 
= 1 × 1 + 4 × 2 + 2 × 1 1 × (-1) + 4 × 3 + 2 × ( -2) 1 × 0 + 4 × ( -1) + 2 × 3 
 2 1 -3
 2 -1 3 0  
-1 0 5 
 
3 × 1 + 0 × 2 + 6 × 1 3 × (-1) + 0 × 3 + 6 × (-2) 3 × 0 + 0 × (-1) + 6 × 3
= 
AA - 1 0 4 2   
  3 4 2 
 3 -4 3 
t
 -3 5 2 -1  0 -2 -1
= 11 7 2  
 
(1)
 9 -15 18  4 + 1 + 9 + 0 2 + 0 + 12 + 0 -6 - 5 + 6 + 0 
=  2 + 0 + 12 + 0 1 + 0 + 16 + 4 -3 + 0 + 8 + 2 
 
1 -1 0   2 0 1   -6 - 5 + 6 + 0 -3 + 0 + 8 + 2 9 + 25 + 4 + 1 
=BA  2 3 -1 1 4 2 
   14 14 -5
1 -2 3   3 0 6 
= 14 21 7 
 
 -5 7 39 
20 21
 2 1 -3
 -1 0 5   2 -1 3 0 
= =  -  ( ) 
 1 0 4 2
 which is A,
 2 -1 3 0 
3 4 2 7. If A - 1 0 4 2  then ind AAt and AtA.
t t
=
t
A , so A
 -3 5 2 -1
As
   
 0 -2 -1  -3 5 2 -1
That is, (At)t = A. (Note that this rule holds for any matrix A.)
8. Solve the following matrix equations for X:
 2 3 -2   2 -3 1 
Exercise 3.2
== - 2A = B if A    5 4 -1 ,
 -1 1 5   
i) 3X and B
 1 -1 2   3 -1 0 
1. If A = [aij]3%4, then show that
== - 3A = B if A   4 2 1
 -2 4 5 
i) I3A = A ii) AI4 = A
 
ii) 2X and B
2. Find the inverses of the following matrices.
 3 -1  -2 3  2i i   2 1
9. Solve the following matrix equations for A:
i)   ii)   iii)   iv)  
2 1   -4 5  i -i   6 3 4 3  2 3   -1 -4   3 1   -1 2   2 0 
(i)   A- = 3 6  (ii) A  -  3 1 =
 
3. Solve the following system of linear equations.  2 2   -1 -2     4 2     -1 5 
2=x1 - 3 x2 5 4=x1 + 3 x2 5 3 x1 - 5 y =
1 
  
3.6 Determinants
=
5 x1 + x2 4  =
3 x1 - x2 7  -2 x + y = -3
i) ii) iii)
 1 -1 2   2 1 -1  1 3 -2 
4. If A  3 2- 5  , B
=
=  1 3 4  and C =  1 2 0  , then ind
     
The determinants of square matrices of order n83, can be written by following
 -1 0 4   -1 2 1   3 4 -1 the same pattern as already discussed. For example, if n = 4
A-B B-A iii) (A - B) - C iv) A - (B - C)

 a11 a12 a14 
i) ii)
a a24 
a13 a11 a12 a13 a14
i 2i   -i 1  2i -1 A =  21 22  , then the determinat of = =
=
5. If A =   2i i   -i i  , then show that
a a23 a21 a22 a23 a24
 a31 a32 a34 
1 -i     
, B and C = A A
 
a33 a31 a32 a33 a34
 a41 a42 a44 
(A + B)C = AC + BC
a43 a41 a42 a43 a44
i) (AB)C = A(BC) ii)
Now our aim is to compute the determinants of various orders. But before describing
6. If A and B are square matrices of the same order, then explain why in general; a method for computation o f determinants of order n83, we introduce the following
i) (A + B)2 ≠ A2 + 2AB + B2 ii) (A - B)2 ≠ A2 - 2AB + B2 deinitions.
iii) (A + B )(A - B ) ≠ A2 - B2
22 23
 a11 a13  a1 j  a1n 

a a23  a2 j  a2 n 
3.6.1 Minor and Cofactor of an Element of a Matrix or its Determinant a12
 21 a22
 a31 a33  a3 j  a3n 
 
a32
A=       , then
 ai1 ai 3  aij  ain 
Minor of an Element: Let us consider a square matrix A of order 3 .Then the minor of If
an element aij, denoted by .Mij is the determinant of the (3 - 1) % (3 - 1) matrix formed by  
ai 2
      
a an 3  anj  ann 
 n1
deleting the ith row and the jth column of A(or|A|). an 2
For example, if
|A| = ai1Ai1 + ai2Ai2 + ai3Ai3 + .... + ainAin for i = 1, 2, 3,...., n
 a11 a12a13 
or |A| = a1j A1j + a2j A2j + a3j A3j + .... + anj Anj for j = 1, 2, 3,...., n
A =  a21 a22a23  , then the matrix obtained by deleting the irst row and the second
Putting i = 1, we have
 
 a31 a32
a33 
|A| = a11 A11 + a12 A12 + a13 A13 + .... + a1n A1n which is called the expansion of |A|by the irst
-------------------
 a11 a12 a13 
--------------------- row.
 a21 a23 
column of A is   (see adjoining igure)  a21 a22 a23  and its determinant is the
   a11 a12 a13 
 a31 a33   a31 a32 a33 
If A is a matrix of order 3, that is, A =  a21 a22 a23  , then:
 
 a31 a32 a33 
minor of an, that is,
M 12 =
a21 a23
a31 a33 |A| = ai1Ai1 + ai2Ai2 + ai3Ai3 + .... + ainAin for i = 1, 2, 3 (1)
or |A| = a1j A1j + a2j A2j + a3j A3j + .... + anj Anj for j = 1, 2, 3 (2)
Cofactor of an Element: The cofactor of an element aij denoted by Aij is deined by Aij = (-1) For example, for i = 1, j = 1 and j = 2, we have
i+j
%Mij |A| = a11 A11 + a12 A12 + a13 A13 (i)
where Mij is the minor of the element aij of A or |A|. or |A| = a11 A11 + a21 A21 + a31 A31 (ii)
or |A| = a12 A12 + a22 A22 + a32 A32 (iii)
= - 21 23
a21 a23 a a
For example, A12 = (-1)1+2 M12 = (-1)3 (iii) can be written as:|A| = a12(-1)1+2 M12 + a22(-1)2+2 M22 + a32(-1)3+2 M32
i.e., |A| = -a12M12 + a22M22 - a32M32
a31 a33 a31 a33
(iv)
Similarly (i) can be written as |A| = a11M11 - a12M12 - a13M13 (v)
3.6.2 Determinant of a Square Matrix of Order n83: Putting the values of M11.M12: and M13 in (iv). we obtain
A = a11 - a12 + a13

a22 a23 a21 a23 a21 a22
The determinant of a square matrix of order n is the sum of the products of each a32 a33 a31 a33 a31 a32
or |A| = a11(a22a33 - a23a32) - a12(a21a33 - a23a31) + a13(a21a32 - a22a31)

elem ent of row (or column) and its cofactor.
(vi)
or |A| = a11a22a33 + a12a23a31 + a13a21a32 - a11a23a32- a12a21a33- a13a22a31 (vi)’
24 25
 1 -2 3 
The second scripts of positive terms are in circular order of
Example 2: Find the cofactors A12, A22 and A32 if A =  -2 3 1  and

anti-clockwise direction i.e., these are as 123, 231, 312 (adjoining igure)
 
 4 -3 2 
while the second scripts of negative terms are such as 132, 213, 321.
ind |A|.
Solution : We irst ind M12, M22 and M32,
An alternative way to remember the expansion of the determinant |A| given in (vi)’ is
-2 1
M 12 = =-4-4=-8; M 22 = = 2 -1 2 = -10 and M 32 =
shown in the igure below. 1 3 1 3
-2 1
= 1- (-6) =7
4 2 4 2
Thus A12 = (-1)1+2 M12 = (-1)(-8 ) = 8 ; A22 = (-1)2+2 M22 = 1(-10)= -10
A32 = (-1)3+2 M32 = (-1)(7 ) = -7 ;
and |A| = a12A12 + a22 A22 + a32 A32 = (-2)8 + 3(-10) + (-3)(-7)
= -16 - 30 + 21 = -25
Note that a11A12 + a21 A22 + a31 A32 = 1(8) + (-2)(-10) + 4(-7)
= 8 + 20 - 28 = 0
and a13A12 + a23 A22 + a33 A32 = 3(8) + 1(-10) + 2(-7)
= 24 - 10 - 14 = 0
 1 -2 3 
Example 1: Evaluate the determinant of A =  -2 3 1 
Similarly we can show that a11A13 + a21 A23 + a31 A33 = 0;
 
 4 -3 2 
a11A21 + a12 A22 + a13 A23 = 0; and a11A31 + a12 A32 + a13 A33 = 0;
3.7 Properties of Determinants which Help in their Evaluation

1 -2 3
Solution : A = -2 3 1 1. For a square matrix A, |A| =IAtI
4 -3 2 2. If in a square matrix A, two rows or two columns are interchanged, the determinant of the
resulting matrix is -|A|.
Using the result (v) of the Art.3.6.2, that is, 3. If a square matrix A has two identical rows or two identical columns, then |A| = 0 .
|A| = a11 M11 - a12M12 + a13M13 , we get, 4. If all the entries of a row (or a column) of a square matrix A are zero, then |A| = 0 .
-2 1 -2 3
5. If the entries of a row (or a column) in a square matrix A are multiplied by a number k U
= - (-2) +3
3 1
-3 2 4 -3
A 1 _, then the determinant of the resulting matrix is k |A|.
4 2 6. If each entry of a row (or a column) of a square matrix consists of two terms, then its
= 1[6 - 1(-3)] + 2[(-2).2 -1.4] + 3[(-2)(-3) -12]

determinant can be written as the sum of two determinants, i.e., if
 a11 + b11 a12 a13 
= (6 + 3) + 2(-4 - 4) + 3(6 -12) =B  a21 + b21 a22 a23  ,
  then
= 9 -16 -18 = -25  a31 + b31 a32 a33 
26 27
a11 + b11 a12 a13 a11 a12 a13 b11 a12 a13 a b c
+B=
a21 b21 = a22 a23 a21 +a22 a23 b21 a22 a23 a b c = 0, (it can be proved by expanding the determinant)
a31 + b31 a32 a33 a31 a32 a33 b31 a32 a33 x y z
7. If to each entry of a row (or a column) of a square matrix A is added a non-zero multiple
of the corresponding entry of another row (or column), then the determinant of the Property 5: If any row (column) of a determinant is multiplied by a non-zero number k, the
resulting matrix is |A|. value of the new determinant becomes equal to k times the value of original determinant.
8. If a matrix is in triangular form, then the value of its determinant is the product of the For example,
entries on its main diagonal.
A=
a11 a12
Now we prove the above mentioned properties of determinants. , multiplying irst row by a non-zero number k, we get
a21 a22
Proporty 1: If the rows and columns of a determinant are interchanged, then the value
= ka11a22 - ka12a21 = k (a11a22 - a12a21) = k 11 12

of the determinant does not change. For example. ka11 ka12 a a
a21 a22 a21 a22
= a11a22 - a12a21 = a11a22 - a21a12 = 11

a11 a12 a a21
(rows and columns are interchanged)
a21 a22 a12 a22 Property 6: If any row (column) of a determinant consists of two terms, it can be written as
the sum of two determinants as given below:
Property 2: The value of a determinant changes sign if any two rows (columns) are
a11 + b11
interchanged. For example,
a12 a13 a11 a12 a13 b11 a12 a13
a21 + b21 a23 = a21 a23 + b21
= a11a22 - a12a21
a11 a12 a22 a22 a22 a23 (proof is left for the reader)
a21 a22 a31 + b31 a32 a33 a31 a32 a33 b31 a32 a33
= a12a21 - a11a22 = -(a11a22 - a12a21) (columns are interchanged)

a12 a11 Property 7: If any row (column) o f a determinant is multiplied by a non-zero number k and
and
a22 a21 the result is added to the corresponding entries of another row (column), the value of the
Property 3: If all the entries in any row (column) are zero, the value of the determinant is determinant does not change. For example,
a11 a12 a11 a12 + ka11

zero. For example,
=
a21 a22 a21 a22 + ka21
(k multiple of C1 is added to C2)
0 a12 a13
a23 = 0 22 -0 +0 = 0 (expanding by C )
a a23 a12 a13 a12 a13 It can be proved by expanding both the sides. Proof is left for the reader.
0 a22
 2 -2 3 4 
a32 a33 a32 a33 a22 a23
 3 1 5 -1
0 a32 a33
A=  , evaluate |A|

 -5 -3 1 0 
Example 3: If
 
Property 4: If any two rows (columns) of a determinant are identical, the value of the
 1 -1 0 2 
determinant is zero. For example,
28 29
1 1 b+c
2 -2 3 4
= x(a + b + c) 1 1 c + a , (by property 5)
3 1 5 -1
A=
Solution :
1 1 a+b
-5 -3 1 0
1 -1 0 2
= x(a + b + c).0 (a C1 and C2 are identical or by property 3)
0 0 3 0
5 -7
=0
=
x 0 1 1
0 1 -1 -1
0 4
0 -8 =0
By R1 + (-2 )R4, R2 + (-3 )R4 and R3 + 5R4
1 -2 3 4
1 10
1 -1
Example 5: Solve the equation
-2 x 1 -1
0 2
Expanding by irst column, we have

Solution: By C3 + C2 and C4 + C2, we have
|A| = 0.A11 + 0.A21 + 0.A31 + 1.A41
0 3 0 0 3 0
=(-1) 4+1 × 4 5 -7 =(-1) 4 5 -7
x 0 1 1
=0
-8 1 10 -8 1 10
0 1 0 0
1 -2 1 2
= ( -1)(- 3)[4 % 10 - (-7)(-8)] = 3(40 - 56) = - 48 -2 x x + 1 x - 1
x a+ x b+c
Example 4: Without expansion, show that x b + x c + a =
x 1 1
2 = 0 (a (-1)2+2 = 1)
0
x c+ x a+b
Expanding by R2, we get 1 1
-2 x + 1 x - 1
Solution : Multiplying each entry of C1 by -1 and adding to the corresponding entry of C2 i.e.,
x 1 1
2 =0
by C2 + (-1)C1, we get
By R3 + 2R2, we get 1 1
0 x+3 x+3
x a + x b + c x a + x + (-1) x b + c
x b + x c +=a x b + x + (-1) x c + a
x 1 1
( x + 3) 1 1 2 =
x c + x a + b x c + x + (-1) x a + b
or 0 (by taking x + 3 common from R3)
x a b+c 1 a b+c  by property 5 or 

0 1 1
= x b c += x1 b c+a  taking x common 

x 1 1
  ⇒ 1 1 2 =0
 
a
x c a+b 1 c a+b  
x+3=0 or
from C1
1 a + (b + c) b + c
0 1 1
 adding the entries of C3 to the 

= x 1 b + (c + a ) c + a ,   ⇒ x = -3
 corresponding entries of C2 
or x = 0 (a R1 and R3 are identical if x = 0)
1 c + ( a + b) a + b Thus the solution set is {-3, 0}.
30 31
3.8 Adjoint and Inverse of a Square Matrix of Order n83

A33 =-
( 1)3+3 =1.(2 - 0) =2
1 0
 a11 a12 a13   A11 A13 

0 2
If A =  a21 a22 a23  , then the matrix of co-factors of A =  A21 A23  ,  3 1 -2 

A12
 A11 A13 
    A A23   2 1- 1 
A12
[ Aij ]3×3 = =-
A22
 a31 a32 a33   A31 A33  Thus
 21
A22
  
 A31 A33   -4 -1 2 
A32
 A11 A21 A31 
and adj A =  A12 A22 A32  ,
A32
 3 -2 -4 
   1 -1 -1
 A13 A23 A33  = [ Aij′ ]3×=
 
and adj A (a A’ij = Aji for i, j=1, 2, 3 )
 -2 1 2 
3
Inverse of a Square Matrix of Order n83: Let A be a non singular

square matrix of order n. If there exists a matrix B such that Since |A|= a11A11 + a12A12 + a13A13
AB = BA = In, then B is called the multiplicative inverse of A and is denoted by A-1. It is obvious = 1(3) + 0(1) + 2(-2)
= 3 + 0 - 4 = -1,
 3 -2 -4   -3 2 4 
that the order of A-1 is n % n.
A -1 = adj A =  1 -1 -1 = -1 1 1 
Thus AA-1 = In and A-1A = In.
-1    
So 1 1
 -2 1 2   2 -1 -2 
If A is a non singular matrix, then
A
A-1 =
 -1 2 
1
adjA
 1 3
Example 7: If A =  1 4  and
A
   2 1 then verify that
1 0 2   2 -1  
Example 6: Find A-1 if A = 0 2 1 
 
1 -1 1 
(AB)t = BtAt
 -1 2   -1 - 4 -3 + 2   -5 -1
 
Solution: We irst ind the cofactors of the elements of A.
AB = 1 4= =  1- 8 3+ 4  =  -7 7  ,
   -2 1    
1 3
Solution: so
A11 =-
( 1)1+1
2 1
=1.(2 + 1) =3, A12 =(-1)1+ 2
0 1
=(-1)(-1) =1  2 -1  2 + 2 6 - 1   4 5 
-1 1
 -5 -7 4 
1 1
( AB )t =  
A13 =
(-1)1+3 =
1.(0 - 2) =
-2, A21 =-
( 1) 2+1 =-
( 1)(0 + 2) =-2  -1 7 5 
0 2 0 2
1 -1 -1 1
1 -2   -1 1 2    -1 1 2  1 -2  
B t At =     A =   and B t =  
A22 =
(-1) =
1.(1 - 2) =
-1, A23 = (-1) = (-1)(-1 - 0) = 1 3 1   2 4 -1   2 4 -1 3 1  
t
and
2+ 2 1 2 2+3 1 0
1 -1
 -1 - 4 1 - 8 2 + 2   -5 -7 4 
1 1
=
=   -1 7 5 
A31 =
(-1)3+1 =
1.(0 - 4) =
-4, A32 =-
( 1)3+ 2 =-
( 1)(1 - 0) =-1  -3 + 2 3 + 4 6 - 1   
0 2 1 2
2 1 0 1
Thus (AB)t = Bt At
32 33
φ φ
+ Determinants
+
3. Matrices and Determinants eLearn.Punjab 3. Matrices and eLearn.Punjab
+ + = + + -
c a+b c+a
Exercise 3.3 a+λ b c
ix) a b+λ c =λ 2 (a + b + c + λ )
Evaluate the following determinants. c+λ
-2 -4 -3 2 -3
a b
5 5 2 1
3 -1 -3 3 -1 1 iii) -1 3
1 1 1
c =(a - b)(b - c)(c - a)
1. i) ii) 4
-2 1 -2 1 -2 -2 5
x) a b
2 6
a2 b2 c2
a+l a-l 1 2 -2
b + c a a2
a 2a a a
a+l a-l v) -1 1 -3
c + a b=b 2 (a + b + c)(a - b)(b - c)(c - a)
iv) a vi) b 2b b
a -l a+l 2 4 -1
xi)
a + b c c2
a c c 2c
2. Without expansion show that

2 3 -1
1 2 -3  5 -2 5 
6 7 8 1 2 3
=
= =

4. If If A =  = 3
2 0 and B- 1 4  ,then find;
-
i) 3 4 5 0 ii) 1 1 0 0 iii) 4 5 6 0
2 -3 5    
0
 -2 -2 1   -2 1 -2 
2 3 4 7 8 9
3. Show that
a13 + a13 a11 a12 a13 a11 a12 a13
i) A12, A22, A32 and |A| ii) B21, B22, B23 and|B|
a11 a12
i) a21 a22 a=23 + a 23 a21 a22 a23 + a21 a22 a 23
a33 + a 33 a31 a32 a33 a31 a32 a 33
5. Without expansion verify that
a31 a32
2 3 0 2 1 0 a+l a a a
3 9 6= a l+ l=
(3 a l + a b +g 1
2
1 a2
ii) 91 1 2 iii) a a ) bc
a+l
1 2 3x
i) b =
g +a = =
2 15 1 2 5 1 a a b
1 0 ii) 2 3 6x 0 iii) 1 b 2 0
b+c g a +b 1
ca
1 1 1 1 1 1 a a 3 5 9x
c
= x= y +z
1 c2
iv) x y z v) b c a b 4abc ab
yz zx xy x 2 y 2 z 2 c c a+b a -b b-c c-a
b -1 a r cos φ 1 - sin φ iv) b-c c-a a -b =
0
vi) a b 0= a 3 b3 vii) 0 1 0+ r = c-a a -b b-c
1 a b r sin φ 0 cos φ bc ca ab
a b+c a+b 1 1 1
=0
viii) b c + a b + c = a 3 + b3 + c3 - 3abc
v)
a b c
c a+b c+a a b c
a+λ mn l l2 1 l2 l3
=
b c version: 1.1 version: 1.1
+λ λ
= λ
+λ 34 35
=
a b c
10. If A is a square matrix of order 3, then show that |KA|= k3|A|.
11. Find the value of λ if A and B singular.
2 2 3
mn l l 1 l l
nl m m 2 = 1 m 2
5 0
m3
 4 λ 3
vi)
8 1
1 2
 
A = 7 3 6 , B= 
lm n n 2 1 n 2 n3
 
2 5
3 1
 2 3 1   
2a 2b2c
vii) a + b b+c
2 0
2b 2 λ -1 3
a + c b + c 2c
7 2 -1 -a 0
12. Which of the following matrices are singular and which of them are
7 2 6 7 2 7 c
viii) 6 3 2= 6 3 5 + 6 3 -3 ix) a -b
non singular?
1 1 2 -1
0
-3 5 1 -3 5 -3 -3 5 4 b -c 0
1 0 3   2 3 -1 1 2 -1 -3
  
i) 3 1 -1 ii) 1 1 0  iii)  
    2 3 1 2 
0 2 4   2 -3 5 
6. Find values of x if
1 x -1 3  
3 1 x 1 2 1  3 -1 3 4 
-1 3 -4 = - 1+ x 1= 0 =
2 1 0
i) 30 ii) 2 0 iii) 2 x 2
-2 x
13. Find the inverse of A = 1 1 0  and show that A-1 A = I3
x 1 0 2 3 6 x
 
7. Evaluate the following determinants:  2 -3 5 
3 4 2 7 2 1 -1
3 -3 9 1 1
0 3 -1 2
14. Verify that (AB)-1 = B-1A-1 if
2 5 0 3 4 0
2 1
1 2 -3 5 2 -1 6 9 7 -1 1  1 2  -3 1  5 1   4 3
i) ii) iii)
=
=
i) A = =
5
4 1 -2 3 -7 2 -2 -2 0 1 -1  -1 0   4 -1 2 2 , B  2 1
       
6 , B ii) A
x 1 1 1 15. Verify that (AB)t = Bt At and if

=
(x + 3)(x - 1)3
1 1 
1 x 1 1
1 -1 2 
8. Show that
=
= 3 2 
1 1 x 1
A   and B
0 3 1   
0 -1
1 1 1 x
9. Find |AAt| and |AtA| if

 2 -1  2 -1
3 4 16. If A =   ( A=-  that ( A-1 )t =( At ) -1
2 1 3 1  3 1 
verify
 3 2 -1  
=
=i) A   ii) A 1
  1
 
2 1 3 17. If A and B are non-singular matrices, then show that
2 3 i) (AB)-1 = B-1A-1 ii) (A-1)-1 = A
36 37
1 1 2   x   1  1 2 3 
 2 -1 8   y  = 12 
[x z ] 1 -1 5 = [1 12 -3]
3.9 Elementary Row and Column Operations on a Matrix
      
or y
Usually a given system of linear equations is reduced to a simple equivalent system by  3 5 4   z   -3  2 8 4 
X t At = B t
applying in turn a inite number of elementary operations which are stated as below:
that is, AX = B (i) (ii)
1. Interchanging two equations.
2. Multiplying an equation by a non-zero number.
1 1 2   x 1
A=-2 1 8 = , X  y  and=
B 12 
3. Adding a multiple of one equation to another equation.
     
where
Note: The systems of linear equations involving the same variables, are equivalent if they  3 5 4   z   -3
have the same solution.
Corresponding to these three elementary operations, the following elementary row A is called the matrix of coeicients.
operations are applied to matrices to obtain equivalent matrices. Appending a column of constants on the left of A, we get the augmented matrix of the
i) Interchanging two rows given system, that is,
ii) Multiplying a row by a non-zero number
iii) Adding a multiple of one row to another row. 1 1 2  1 
 2 -1 8  12 
 
(Appended column is separated by a dotted line segment)
 3 5 4  -3
Note: Matrices A and B are equivalent if B can be obtained by applying in turn a inite
number of row operations on A.
Now we explain the application of elementary operations on the system-of linear equations
Notations that are used to represent row operations for I to III are given below: and the application of elementary row operations on the augmented matrix of the system
Interchanging Ri and Rj is expressed as RiGRj . writing them side by side.
k times Ri, is denoted by kRi D R’i
Adding k times Rj to Ri is expressed as Ri + kRj D R’i x + y + 2z = 1  1 1 2  1
 2
=
2 x + - y + 8 z 12
-  1 8  12 
(R’i is the new row obtained after applying the row operation).
 
3 x + 5 y + 4 z-=3  3 5 4  -3
For equivalent matrices A and B, we write A R .B.

Adding -2 times the irst equation to the (By R2+ (-2)R1 DR’2 and
If A R B then B R A. Also if A R B and B R C, then A R C. Now we state the elementary column
    
second and -3 times the irst equation to R3 + (-3)R1D R’3, we get)
operations and notations that are used for them.
i) Interchanging two columns Ci G Cj
ii) Multiplying a column by a non-zero number kCi D C’i the third, we get
iii) Adding a multiple of one column to another column Ci + kCj D C’i
x + y + 2z = 1  1 1 2  1

R 0 3 4  10 
Consider the system of linear equations;
x + y + 2z = 1  =
- 3 y + 4 z 10 -
  
2 x - y = 8 z = 12  which can be written in matrix forms as 2 y - 2 z-=6  0 2 -2  -6 
3 x + 5 y + 4 z-=3
38 39
1 2 3  1 0 0 
 1 -1 5   1 -3 2 
Interchanging the second and third equations, we have (By R2 GR3, we get)
   
x + y + 2z = 1 1 1 2  1  2 8 4  C  2 4 -2  By C2 + (-2)C1 D C’2 and
    
2 y - 2 z-=6  R 0 2 -2  -6     
 
C3 + (-3)C1 D C’3
     
  1 12 -3  1 10 -6 
=
- 3 y + 4 z 10 -
 0 3 4  10 
1 0 0  1 0 0 
 1 2 -3   1 1 -3
   
1 1
Multiplying the second equation by , we get By R2 DR’2, we get.
C  2 -2 4  By C2 ↔ C3 C  2 -1 4  By C2 → C'2
2 2 1
   
x + y + 2z =1 1 1 2  1       
2

R 0 1 -1  -3
  
y - z =-3   1 -6 10   1 -3 10 
 
10 
-3 y + 4 z = 0 -3 4  10 
1 0 0
By R3 + 3R2DR’3, we obtain, the third, we obtain, 1 1 0
 
Adding 3 times the second equation to
C  2 -1 1  By C3 + 3C2 → C'3
x + y + 2z =1 1 1 2  1   
   
y - z =-3 R 0 1 -1  -3  1 -3 1 
 
z = 1  0 0 1  1 
1 0 0 
[x z ] 1 1 0  = [1 -3 1]
The given system is reduced to the triangular form which is so called because on the left
 
Thus y
 2 -1 1 
the coeicients (of the terms) within the dotted triangle are zero.
Putting z = 1 in y - z = -3 , we have y - 1 = - 3 ⇒ y = - 2
Substiliting z = 1, y = -2 in the irst equation, we get
x + (-2 ) + 2(1) = 1 ⇒ x = 1
or [ x + y + 2z y-z z ] = [1 -3 1]
Thus the solution set o f the given system is {(1, -2,1)}.
x + y + 2z =1

Appending a row of constants below the matrix At, we obtain the
⇒ y - z =-3
1 2 3  z = 1 
 
augmented matrix for the matrix equation (ii), that is  1 -1 5 
2 8 4 
Upper Triangular Matrix: A square matrix A = [aij] is called upper triangular if all elements
  below the principal diagonal are zero, that is,
   
 1 12 -3
aij = 0 for all i > j
Lower Triangular Matrix: A square matrix A = [aij] is said to be lower triangular if all elements
Now we apply elementary column operations to this augmented matrix. above the principal diagonal are zero, that is,
aij = 0 for all i < j
40 41
Triangular Matrix: A square matrix A is named as triangular whether it is upper triangular For diagonal elements j = i, so
or lower triangular. For example, the matrices aii = -aii or 2aii = 0 ⇒ aii = 0 for i = 1, 2, 3, ......., n
 0 -4 1 
1 0 0 0
1 2 3   For example if B =  4 0 -3 ,then
0 1 4  and  3 2 0 0  
 are triangular matrices of order 3 and 4  -1 3 0 
  4 1 5 0
0 0 6   
 -1 2 3 1
 0 4 -1  0 -4 1 

B =-4 0 3 =  (-1)  4 0 -3 = -B
   
respectively. The irst matrix is upper triangular while the second is lower triangular. t
Note: Diagonal matrices are both upper triangular and lower triangular.  1 -3 0   -1 3 0 
Symmetric Matrix: A square matrices A = [aij]n x n is called symmetric if At = A. Thus the matrix B is skew-symmetric.
From At = A, it follows that [a’ij]n x n = [aij]n x n Let A = [aij] be an n x m matrix with complex entries, Then the n x m matrix [a ij] where a ij is the
which implies that a’ij = aji for i, j = 1, 2, 3, ......., n. complex conjugate of aij for all i, j, is called conjugate of A and is denoted by A . For example, if
but by the deinition of transpose, a’ij = aji for i, j = 1, 2, 3, ......., n.
3 - i -i  3 - i -i  3 + i i 
==
A=
Thus aij = aji for i, j = 1, 2, 3, ......., n.
  , then A    
 2i 1 + i   2i 1 + i   -2i 1 - i 
and we conclude that a square matrix A = [aij]n x n is symmetric if aij = aji.
For example, the matrices
Hermitian Matrix: A square matrix A = [aij]nxn with complex entries, is called hermitian if ( A )
t
1 -1
a g 3 0 5 6
From, ( A ) = A it follows that  aij′  = [aij]nxn which implies that a ij/ = aij for i, j = 1 ,2 ,3,....,n but
3 2 = A.
1 3  h 
f and   are symmetric.
h
3 2  ,   2 5 1 -2 
t
 
n×n
b
 g c   
 -1 6 -2 3 
f by the deinition of transpose, a ij/ = aij for i, j = 1, 2, 3,......, n.
Thus aij = aij for i, j = 1, 2, 3,......, n and we can say that a square matrix
Skew Symmetric Matrix : A square matrix A = [aij]n x n is called skew symmetric or anti-
symmetric if At = -A.
A = [aij]nxn is hermitian if aij = a ji for i, j = 1, 2, 3, .... ,n.
From At = -A, it follows that [a’ij] = for i, j = 1, 2, 3, ......., n

For diagnal elements, j = i so aii = a ii which implies that aii is real for i = 1, 2, 3, ..., n
which implies that a’ij = -aij for i, j = 1, 2, 3, ......., n

 1 1 - i
For example, if A =   , then
1 + i 2 
but by the deinition of transpose a’ij = aji for i, j = 1, 2, 3,..., n
Thus -aij = aji or aij = -aji
⇒ ( A) = 
 1 1 + i  1 1 - i
A=    =A
- +
t
Alternatively we can say that a square matrix A = [aij]nxn is anti-symmetric 1 i 2  1 i 2 

if aij = -aji . Thus A is hermitian.
42 43
hermitian or anti-hermitian if ( A ) = -A.

Skew Hermitian Matrix: A square matrix A = [aij]nxn with complex entries, is called skew- Note: Some authors do not require the condition (iii)
t
From ( A ) = -A, it follows that  a ij/  = [-aij]nxn

0 1 -2 4  1 2 -3 4 
The matrices 0 0 1 2 and 0
 0 1 2  are in echelon form
t
   
n×n
which implies that a ij/ = -aij for i, j = 1, 2, 3, ..., n. 0 0 0 0  0 0 0 1 

but by the deinition of transpose, a ij/ = a ji for i, j = 1, 2, 3, ...., n. 0 0 1 2 0 1 -2 
Thus -aij = a ji or aij = -a ji , for i, j = 1, 2, 3, ...., n. but the matrices 0 1 3 -1 and 0 0 -1 are not in echelon form.
   
and we can conclude that a square matrix A = [aij]nxn is anti-hermitian if aij = -a ji 0 0 0 0  0 0 4 
For diagonal elements j = i, so aii = -a ii ⇒ aii + -a ii = 0

which holds if aii = 0 or aii = iλ where λ is real
Reduced Echelon Form of a Matrix: An m x n matrix A is said to be in reduced (row)
because 0 + 0 = 0 or iλ + iλ = iλ - iλ = 0
echelon form if it is in (row) echelon form and if the irst non-zero entry (or leading entry) in
Ri lies in Cj , then all other entries of Cj are zero.
 0 2 + 3i 
For example, if A =   , then
 - +  0 1 0 4  1 2 0 0 
The matrices 0 0 1 2  and 0 0 1 0  are in (row) reduced
2 3i 0
 0 2 + 3i     
A= 0 0 0 0  0 0 0 1 
 -2 + 3i 0 
( A)  0 -2 + 3i   0 2 - 3i 
echelon form.
⇒ =   = (-1)  -2 - 3i  = -A
+
t
 2 3i 0   0  Example 1: Reduce the following matrix to (row) echelon and reduced (row) echelon form,
 2 3 -1 9 
1 -1 2 -3
Thus A is skew-hermitian.
 
3.10 Echelon and Reduced Echelon Forms of Matrices  3 1 3 2 
 2 3 -1 9  1 -1 2 -3
1 -1 2 -3 R 2 3 -1 9  By R1 ↔ R2
   
Solution:
In any non-zero row of a matrix, the irst non-zero entry is called the leading entry of  3 1 3 2   3 1 3 2 
that row. The zeros before the leading entry of a row are named as the leading zero entries 1 -1 2 -3 1 -1 2 -3
  By R2 +( - 2)R1 → R '2
R 0 5 -5 15 R 0 1 -1 3  By R2 → R '2
  and R +( - 3)R → R '   
of the row. 1
0 4 -3 11  0 4 -3 11 
Echelon Form of a Matrix: An m x n matrix A is called in (row) echelon form if 3 1 3 5
1 -1 2 -3 1 0 1 0
i) In each successive non-zero row, the number of zeros before the leading entry is greater
R 0 1 -1 3  By R3 +( - 4)R2 → R '3 R 0 1 -1 3  By R1 +1.R2 → R '1

than the number of such zeros in the preceding row,
   
0 0 -1 -1 0 0 1 -1
ii) Every non-zero row in A precedes every zero row (if any),
iii) The irst non-zero entry (or leading entry) in each row is 1.
44 45
Interchanging R1 and R3 we get..

1 2 -2  0 0 1  1 2 -2  0 0 1 
1 0 0 1  3 4 2  0 1 By R2 + ( - 3)R1 → R'2
By R1 +( - 1)R3 → R '1 0  R 0 -2 8  0 1 -3
R 0 1 0 2      and R + ( - 2) R → R'
  and R + 1.R → R '  2 5 -1  1 0 0  0 1 3  1 0 -2 
0 0 1 -1
3 1 3
2 3 2
1 -1 2 3  1 0 0 1  By - R2 → R'2 ,
1
Thus 0 1 -1 3  and 0 1 0 2  are (row) echelon and reduced (row) echelon forms
we get
     
2
0 0 1 -1 0 0 1 -1 1 2 -2  0 0 1  1 0 6  0 0 -2 

  
3  3  By R3 + ( - 1)R2 → R'3
0 1 -4  0 -  R 0 1 -4  0 -
1 1
 2  2  and R1 + ( - 2) R2 → R'1
of the given matrix respectively.
0 1 3  1 0 -2   7
Let A be a non-singular matrix. If the application of elementary row operations on A I in 2 2
succession reduces A to I, then the resulting matrix is I  A-1 .  0 0 7  1
1
- 
 2 2
 A R3 → R'3 , we have
1
Similarly if the application of elementary column operations on...in succession reduces
By
 I 
7
  1 0 0  - 6 
1 -2  
1 
1 0 6  0
4
   
I
1  By R1 + ( - 6)R3 → R'1
A to I, then the resulting matrix is -1 7 7
0 1 -4  0 1 3  
- - -
A
2 2   2  and R2 + 4R3 → R'2
4 3

R 0 1 0 
 A  I  
Thus A I R I  A and... C .... 
  1 
7 14
-  0 0 1 
-1
 I    A-1  0 0 1  -
2   2 
1 1 1 1 1

   7 14 7 14
 6 
 2 5 -1 - 7 1 
4
Example 2: Find the inverse of the matrix A =  3 4 2   

 
7

Thus the inverse of A is - - -
1
1 2 -2   7 2
4 3
 2 5 -1  1 1
14
 - 
Solution: A =  3 4 2  = 2(-8 - 4) - 5(-6 - 2) -1(6 - 4) = -24 + 40 - 2  7 2 
1
 
14
1 2 -2  Appending I3 below the matrices A, we have
 2 5 -1
= 40 - 26 = 14 As |A| ≠ 0, so A is non-singular. 3 4 2 
 
1 2 -2 
 2 5 -1  1 0 0   
1 0 0 
Appending I3 on the left of the matrix A, we have  3 4 2  0 1 0   
 
1 2 -2  0 0 1  0 1 0 
0 0 1 
 
46 47
Interchanging C1 and C3, we get
 2 5 -1  -1 5 2   1 5 2
3 4 2   2 4 3  -2 4 3 
       6 
1 2 -2   -2 2 1   2 2 1 - 7 1 
4
  C   C   By (-1)C1 D C’1  
7
1 0 0    0 0 1    0 0 1  Thus the inverse of A is  - -
1
       7 2
4 3
     1
 0 1 0 1 
14
0 0 1  1 0 0   -1 0 0   -
0 1 0 0 1 0
       7 2 
1
14
By C2 + (-5)C1 D C’2 and C3 + (-2)C1 D C’3, we have
Rank of a Matrix: Let A be a non-zero matrix. If r is the number of non-zero rows when it is
 
 1 0 
-
reduced to the reduced echelon form, then r is called the (row) rank of the matrix A.
 1 0 0   -2 1 7
0
 -2 14 7   
   - -  1 -1 2 -3 
4
 2 - 7 -3  2 0 7 -7 
 2 -8 -3
2 4 3
.....  C 
7 ..... 
By C2 → C '2  
 1 Example 3: Find the rank of the matrix
 0 0 1  C  0 00 11  B 14 →  3 1 12 -11
.....
   
1 -1 2 -3  1 -1 2 -3
 0 1 0   00 1 00  By R2 + ( -2 ) R1 → R '2
 -1 5 2    Solution:  2 0 7 -7  R 0 2 3 -1
        and R + ( -3) R → R '
-  3 1 12 -11 0 4 6 -2 
14
- 
 1 14 2 
3 1 3
5
1 -1 -3  1 -1 -3 
 1  1
By C1 + (2)C2 D C’1 and C3 + (-7)C2 D C’3 we have 2 2
R 0 2 - By R2 → R '2 R 0 1
 -  By R3 + ( - 4)R2 → R '3
3 1 3
1 0 0   1 0 0   2  2
0 1 0   0 1 0  0 4 -2  0 0 0 
2 2 2
     6  0
6   0 0 1 
 7 - 7 1  ......  By C +  - 6  C → C '  7
4
......   6    3 -
 7  2
7
 0 C -
1    7
1 
1 0
 
1 1
4
  and C +  4  C → C '
2
  3 R 0 1 -  By R1 + 1.R2 → R '1
0
 1 1   4 1
7
7  2
3 1
   - -  0 0 0 0 
1 3 2 2
2
 7 14 2   7 14 2
 
- 2 5 - 1   1 -
1  
 7 14 2   7 2 
1
14 As the number of non-zero rows is 2 when the given matrix is reduced to the reduced
echelon form, therefore, the rank of the given matrix is 2.
48 49
Exercise 3.4 10. Find the rank of the following matrices
 1 -2 5   -3 1 -2  1 -4 -7   3 -1 3 0 -1
If A =  -2 3 -1 and B =  1 0 -1 ,then show that A + B is symmetric. 1 -1 2 1  2
    -5 1  1 2 -1 -3 -2 
i)  2 -6 5 1  ii)    
1.
 5 -1 0   -2 -1 2    1 -2 3 
iii)
2 3 4 2 5 
 3 5 4 -3    
1 2 0 3 -7 4   2 5 -2 -3 3 
If A =  3 2 -1 , show that
 
2.
 -1 3 2 
3.11 System of Linear Equations
i) A + At is symmetric ii) A - At is skew-symmetric.
3. If A is any square matrix of order 3, show that An equation of the form:
i) A + At is symmetric and ii) A - At is skew-symmetric. ax + by = k (i)
4. If the matrices A and B are symmetric and AB = BA, show that AB is symmetric. where a ≠ 0, b ≠ 0, k ≠ 0
5. Show that AAt and AtA are symmetric for any matrix of order 2 x 3. is called a non-homogeneous linear equation in two variables x and y.
i 1 + i 
Two linear equations in the same two variables such as:
6. If A =   ,show that
 -  a1 x + b1 y =
k1 

1 i
a2 x + b2 y =
i) A + ( A ) is hermitian ii) A - ( A ) is skew-hermitian.
k2 
(I)
t t
is called a system of non-homogeneous linear equations in the two variables x and y if

7. If A is symmetric or skew-symmetric, show that A2 is symmetric. constant terms k1, k2 are not both zero.
If in the equation (i), k = 0 , that is, ax + by = 0 , then it is called a
 1 
homogeneous linear equation in x and y.
8. If A = 1 + i  , ind A ( A ) .
If in the system (I), k1 = k2 = 0 , then it is said to be a system of homogenous linear
 
t
 i 
equations in x and y.
An equation of the form:
9. Find the inverses of the following matrices. Also ind their inverses by using row and ax + by + cz = k . . . . (ii)
column operations. is called a non-homogeneous linear equation in three variables x, y and z if a ≠ 0, b ≠ 0, c ≠ 0
and k ≠ 0. Three linear equations in three variables such as:
 1 2 -3 1 2 -1 1 -3 2 
i)  0 -2 0  ii) 0 -1 3  2 1 0 a1 x + b1 y + c1 z =
k1 
      
iii)
 -2 -2 2  1 0 2   0 -1 1  a2 x + b2 y + c2 z =k2 
k3 
(II)
a3 x + b3 y + c3 z =
50 51
is called a system of non-homogeneous linear equations in the three variables x, y and z, if
2 5 -1  5  1 2 -2  -3
constant terms k1, k2 and k3 are not all zero.
If in the equations (ii) k = 0 that is, ax + by + cz = 0. 3 4 2  11  R  3 4 2  11  BY R 1 ↔ R 3
    
1 2 -2  -3  2 5 -1  5 
then it is called a homogeneous linear equation it x, y and z.
If in the system (II), k1 = k2 = k3 = 0, then it is said to be a system of homogeneous linear
1 2 -2  -3 1 2 -2  -3
R 0 -2 8  20 BY R2 + ( - 3)R1 → R2′ R 0 -2 8  20  BY R3 + ( - 2)R1 → R3′

equations in x ; y and z.
   
A system of linear equations is said to be consistent if the system has a unique
solution or it has ininitely many solutions.  2 5 -1  5  0 1 3  11 
BY - R2 → R2′ , we get
A system of linear equations is said to be inconsistent if the system has no solution. 1
The system (II), consists of three equations in three variables so it is called 3 x 3 linear 2
system but a system of the form:
1 2 -2  -3  1 0 6  17 
0 1 By R1 + ( - 2)R2 → R'1
x - y + 2z = 6
 -4  -10  R 0 1 -4  -10 
2 x + y + 3z =4
   and R + ( - 1), R → R'
0 1 3  11  0 0 7  21  3 2 3
1 0 6  17  1 0 0  -1
By R1 + ( - 6)R3 → R'1
R 0 1 -4  -10 BY R3 → R′3 R 0 1 0  2 

is named as 2 x 3 linear system.
    and R + 4 R , R'
Now we solve the following three 3 x 3 linear systems to determine the criterion for a system 1
0 0 1  3  0 0 1  3 
to be consistent or for a system to be inconsistent. 7 2 3 2
Thus the solution is x = -1, y = 2 and z = 3.

2x + 5 y - z = 5 x + y + 2z = 1 
 
3x + 4 y + 2 z =
11  2x - y + 7z = 11 
The augmented matrix for the system (2) is
x + 2 y - 2 z-=3 3 x + 5 y + 4 z-=3
....(1), (2)
x - y + 2z = 1  1 1 2  1 
  2 -1 7  11 
2 x - 6 y + 5z = 7   
 3 5 4  -3
3 x + 5 y + 4 z-=3
and ....(3)
Adding (-2)R1 to R2 and (-3)R1 to R3, we get

The augmented matrix of the system (1) is
1 1 2  1 1 1 2 1
 2 5 -1  5   2 -1 7  11  R 0 -3 3 9

 3 4 2  11    


   3 5 4  -3 0 2 -2  -6 
1 2 -2  -3
1 1 2  1 1 0 4
   BY R1 + ( -1) R2 → R1′
-1  -3
3 
-1  -3 By - R2 → R2′ R 0 1
    and R + ( -2 ) R → R′
We apply the elementary row operations to the above matrix to reduce it to the equivalent 1
R 0 1
0 2 -2  -6  0 0 0  0 
3
reduced (row) echelon form, that is, 3 2 3
52 53
The system (2) is reduced to equivalent system The third equation 0z = 4 has no solution, so the system as a whole has no solution. Thus the
x + 3z = 4 system is inconsistent.
y-z=-3 We see that in the case of the system (1), the (row) rank of the augmented matrix and the
0z = 0 coeicient matrix of the system is the same, that is, 3 which is equal to the number of the
The equation 0z = 0 is satisied by any value of z. variables in the system (1).
From the irst and second equations, we get Thus a linear system is consistent and has a unique solution if the
x = -3z + 4 .......(a) (row) rank of the coeicient matrix is the same as that of the augmented matrix of the
and y=z-3 ........(b) system.
As z is arbitrary, so we can ind ininitely many values of x and y from equation (a) and (b) In the case of the system ( 2), the (row) rank of the coeicient matrix is the same as that
or the system (2), is satisied by x = 4 - 3t, y = t - 3 and z = t for any real value of t. of the augmented matrix of the system but it is 2 which is less than the number of variables
Thus the system (2) has ininitely many solutions and it is consistent. in the system (2).
Thus a system is consistent and has ininitely many solutions if the (row) ranks of the
1 -1 2  1  coeicient matrix and the augmented matrix of the system are equal but the rank is less
The augmented matrix of the system (3) is  2 -6 5  7 
 
than the number of variables in the system.
 3 5 4  -3
In the case of the system (3), we see that the (row) rank of the coeicient matrix is not
equal to the (row) rank of the augmented matrix of the system.
Adding (-2)R1, to R2 and (-3)R1, to R3 we have Thus we conclude that a system is inconsistent if the (row) ranks of the coeicient matrix
and the augmented matrix of the system are diferent.
1 -1 2  1  1 -1 2  1 
 2 -6 5  7  R 0 -4 1  5 
3.11.1 Homogeneous Linear Equations
   
 3 5 4  -3 0 8 -2  -6  Each equation of the system of following linear equations:
 1 0  .( i ) 
a11 x1 + a12 x2 + a13 x3 =

0  .( ii ) 
  - 
1 -1 2  1  a21 x1 + a22 x2 + a23 x3 =
7
1 0
 
0  .( iii ) 
  5  By R1 + 1.R2 → R1′
4 4
R 0 1 - 
 -  By - R2 → R2′ R 0 1 -  - a31 x1 + a32 x2 + a33 x3 =
 4  and R3 + ( -8 ) R2 → R3′
1 5 1 1
 4
0 8 -2  -6  0 0 0  4  is always satisied by x1 = 0, x2 = 0 and x3 = 0, so such a system is always consistent. The
4 4 4
   
 
solution (0, 0, 0) of the above homogeneous equations (i), (ii), and (iii) is called the trivial
solution. Any other solution of equations (i), (ii) and (iii) other than the trivial solution is
Thus the system (3) is reduced to the equivalent system
called a non-trivial
x+ -z=
7 1
solution. The above system can be written as
0 
4 4
y- -z=  
1 5
AX = O , where O = 0 
0z = 4
4 4
0 
54 55
x1 + x2 + x3 = 0

If |A| ≠ 0, then A is non-singular and A-1 exists, that is,
x1 - x2 + 3 x3 = 0  has only the trivial solution,
0 
A-1(AX) = A-1O = O
 x1  0  x1 + 3 x2 - 2 x3 =
(A A )X = O ⇒ X = O , i.e.,  x2  = 0 
 
-1
or
 x  0 
 3
because in this case
In this case the system of homogeneous equations possesses only the trivial solution.
-2 2
Now we consider the case when the system has a non-trivial solution. 1 1 1 1 0 0
A = 1 -1 3 = 1 -2 2 = = 6 - 4= 2 ≠ 0
2 -3
Multiplying the equations (i), (ii) and (iii) by A11, A21 and A31 respectively and adding the
resulting equations (where A11, A21 and A31 are cofactors of the corresponding elements of A), 1 3 -2 1 2 -3
Solving the irst two equations of the above system, we get x1 = -2x3 and x2 = x3. Putting
we have
x1 = -2x3 and x2 = x3 in the expression.

(a11 A11 + a21 A21 + a31 A31)x1 + (a12 A11 + a22 A21 + a32 A31)x2 + (a13 A11 + a23 A21 + a33 A31)x3 = 0, that is,
|A|x1 = 0. Similarly, we can get |A|x2 = 0 and |A|x3 = 0
For a non-trivial solution, at least one of x1, x2 and x3 is diferent from zero. Let x1 ≠ 0 , then x1 + 3x2 - 2x3 , we have - 2x3 + 3(x3) - 2x3 = - x3, that is,
from |A|x1 = 0, we have |A| = 0. the third equation is not satisied by putting x1 = -2x3 and x2 = x3 but it is satisied only if x3 =
For example, the system 0. Thus the above system has only the trivial solution.
x1 + x2 + x3 = ( I) 
3.11.2 Non-Homogeneous Linear Equations
( II ) 
0
x1 - x2 + 3 x3 =
( III )
0
x1 + 3 x2 - x3 =
Now we will solve the systems of non-homogeneous linear equations with help of the
0 following methods.
has a non-trivial solution because i) Using matrices, that is, AX = B ⇒ X = A-1B.
-2 2
1 1 1 1 0 0 ii) Using echelon and reduced echelon forms
A=1 -1 3 =1 -2 2 = =
2 -2
iii) Using Cramer’s rule.
0
1 3 -1 1 2 -2
x1 - 2 x2 + x3 - = 4

Example 1: Use matrices to solve the system 2 x1 - 3 x2 + 2 x3- =6
Solving the irst two equations of the system, we have
2 x1 + 2 x2 + x3 =5 
2x1 + 4x3 = 0 (adding (I) and (II))
⇒ x1 = -2x3
and 2x2 - 2x3 = 0 (subtracting (II) from (I))
⇒
Solution: The matrix form of the given system is
x2 = x3
Putting x1 = -2x3 and x2 = x3 in (III), we see that (-2x3) + 3(x3) - x3= 0, which shows that the
1 -2 1   x1   -4 
equation (I), (II) and (III) are satisied by  2 -3 2-  x  =  6
x1 = -2t, x2 = t and x3 = t for any real value of t.   2  
Thus the system consisting of (I), (II) and (III) has ininitely many solutions. But the system  2 2 1   x3   5 
56 57
or AX = B (i) Example 2: Solve the system;

1 -2 1   x1   -4 
A-=  3= 
- 2 ,X  x2=  and B  6 
      x1 + 3 x2 + 2 x3 =3 

where 2
 2 2 1   x3   5  4 x1 + 5 x2 - 3 x3- =3,
1 -2 1 1 -2 1 42 
3 x1 - 2 x2 + 17 x3 =
As A= 2 -3 2= 0 1 0 By
+ -R2 ( →
2 )R1 R′2
2 2 1 2 2 1 by reducing its augmented matrix to the echelon form and the reduced echelon form.
( -1)
= (1 2 ) =
=- -1 , that is,
2+ 2 1 1 Solution: The augmented matrix of the given system is
2 1
1 3 2  3 
 4 5 -3  -3
|A| ≠ 0, so the inverse of A exists and (i) can be written as
 
X = A-1B (ii)
Now we ind adj A.  3 -2 17  42 
We reduce the above matrix by applying elementary row operations, that is,
 -7 2 10 
 A -11 = 7=
,A12 2= ,A13 10= ,A21 4 
 Aij  =  4 -1 -6  ,
  A- = 1,A32 0 ,A33 1  1 3 2  3  1 3 2  3 
 4 5 -3  -3 R 0 -7 -11  -15 By R2 + ( -4 ) R1 → R2′
 22 - 1=- 23= =
6 ,A31 =
3×3
 -1 0 1 
Since ,A
    and R + ( -3) R → R′
 -7 4 -1  3 -2 17  42  0 -11 11  33 
AA=  2 -1 0 
3 1 3
=
 
So adj
10 -6 1  1 3 2  3 
R 0 -11 11  33  By R2 ↔ R3
 -7 4 -1  7 -4 1   
and A-1 = adjA=  2 -1- 0  =  2 1 0 0 -7 -11  -15
   
1 1
-1
10 -6 1   -10 6 -1
A
1 3 2  3  1 3 2  3 
 x1   -4   7 -4 1   -4   -28 + 24 + 5    1  
R 0 1 -1  -3 By  -  R2 → R2′ R 0 1 -1  -3  By R3 + 7 R2 → R3′
 
Thus  x2  = A-1  -6  = -2 1 0   -6  = 8 - 6 + 0  ,i.e.,
 11 
0 -7 -11  -15 0 0 -18  -36 

        
 x3   5   -10 6 -1  5   40 - 36 - 5 
1 3 2  3 
 1
R 0 1 -1  -3 By  -  R3 → R3′
 x1   1     18 
x  =  2  0 0 1  2 
 2  
 x3   -1
Hence x1= 1, x2 = 2 and x3 = - 1.
The equivalent system in the (row) echelon form is
58 59
x1 + 3 x2 + 2 x3 =
3  These are three linear equations in three variables x1, x2, x3 with coeicients and constant

x2 - x3-= 3 terms in the real ield R. We write the above system of equations in matrix form
x3 = 2  as: AX = B (2)
Substituting x3 = 2 in the second equation gives: x2 - 2 = -3 ⇒ x2 = -1

Putting x2 = -1 and x3 = 2 in the irst equation, we have  x1   b1 
where=A =
a=  x  b 
x1 + 3 (-1) + 2(2) = 3 ⇒ x1 = 3 + 3 - 4 = 2. ij  3×3 , X  2
and B
 2
Thus the solution is x1 = 2, x2 = -1 and x3 = 2  x3  b3 
1 3 2  3
We know that
Now we reduce the matrix 0 1 -1  -3 to reduced (row) echelon form, i.e.,
the matrix equation (2) can be written as: X = A-1B (if A-1 exists)
 
0 0 -1  2  Note: A-1(AX) ≠ BA-1
1 3 2  3  1 0 5  12 
We have already proved that A =
-1
0 1 -1  -3 R 0 1 -1  -3 By R1 + ( -3) R2 → R1′
1
adj A and
  
A
0 0 1  2  0 0 1  2 
 A11 A21 A31 
1 0 0  2 
By R1 + ( -5 ) R3 → R1′
Aij′ 
adj A =
= A
 12
A22 A32 
=
 ( A′ Aji )
R 0 1 0  -1
3×3
 A13 A23 A33 
  andR + 1.R → R′
ij
0 0 1  2 
 x1   A11 A31   b1   A11b1 + A21b2 + A31b3 
2 3 2
1  
Thus  x2=  1 
A32 =b2 
A21
+ +
  A   A 
The equivalent system in the reduced (row) echelon form is A12 A22 A b A b A b
 x3   A13 A33  b3   A13b1 + A23b2 + A33b3 
12 1 22 2 32 3
x1 = 2
x2 = - 1
A23
 A11b1 + A21b2 + A31b3 

 
x3 = 2
 x1   
 x  =  A12b1 + A22b2 + A32b3 
which is the solution o f the given system. A
 2  
 x3   
3.12 Cramer’s Rule
A b +A b +A b 
A
 13 1 
 
23 2 33 3
Consider the system of equations, A
a11 x1 + a12 x2 + a13 x3 =

b1 
b1 a12 a13

a21 x1 + a22 x2 + a23 x3 =
b2 
b2 a22 a23
b1 A11 + b2 A21 + b3 A31
b3  =
=
(1)
a31 x1 + a32 x2 + a33 x3 =
b3 a32 a33
Hence x1 (i)
A A
60 61
a11 b1 a13 3 1 -4
1 -4
3 (1 + 8 ) - 1(1 - 4 ) - ( 2 + 1)
a21 b2 a23 1
b1 A12 + b2 A22 + b3 A32 -1 2 1
=
= =
=
a31 b3 a33 (ii)
x2 x3
A A 9 9
27 + 3 - 12 18
a11 a12 b1 = = = 2
9 9
Hence x1 = -1 , x2 = 1, x3 = 2
a21 a22 b2
b1 A13 + b2 A23 + b3 A33
=
=
a31 a32 b3
x3 (iii)
A A Exercise 3.5
The method of solving the system with the help of results (i), (ii) and (iii) is often referred to
1. Solve the following systems of linear equations by Cramer’s rule.
as Cramer’s Rule.
2x + 2 y + z = 3 2 x1 - x2 + x3 = 5  2 x1 - x2 + x3 =8
3 x1 + x2 - x3 = -4    
 i) 3 x - 2 y - 2 z =1 ii) 4 x1 + 2 x2 + 3 x3 =8 iii) x1 + 2 x2 + 2 x3 =6
Example 3: Use Crammer’s rule to solve the system. x1 + x2 - 2 x3 = -4 
5 x + y - 3z = 2  3 x1 - 4 x2 - x3 = 3  x1 - 2 x2 - x3 =1 
- x1 + 2 x2 - x3 =1 
3 1 -1
Solution: Here A = 1 1 -2 = 3 ( -1 + 4 ) - 1.( -1 - 2 ) - 1.( 2 + 1)
2. Use matrices to solve the following systems:
-1 2 -1 x - 2 y + z =-1 2 x1 + x2 + 3 x3 =
3  x+ y= 2 
  
=9+3-3=9 i) 3 x + y - 2 z =4 x1 + x2 - 2 x3 =0  iii) 2 x - z =1 
1  -4  -1
ii)
y-z= -3x1 - x2 + 2 x3 = 2 y - 3z =
-4 1 -1
-4 1 -2
-4 ( -1 + 4 ) - 1( 4 + 2 ) - 1( -8 - 1)
3. Solve the following systems by reducing their augmented matrices
1 2 -1
=
=
to the echelon form and the reduced echelon forms.
So x1
9 9
-12 - 6 + 9 -9 x1 - 2 x2 - 2 x-3 = 1 x + 2y + z = 2  x1 + 4 x2 + 2 x3 =2 
= = = 1   
i) 2 x1 + 3 x2 + x3 = 1  ii) 2 x + y + 2 z =-1 2 x1 + x2 - 2 x3 =9 
1  9  12 
9 9 iii)
-4 -1 3 5 x1 - 4 x2 - 3 x3 = 2x + 3y - z = 3 x1 + 2 x2 - 2 x3 =
1 -4 -2
3 ( 4 + 2 ) + 4 ( -1 - 2 ) - 1(1 - 4 )
4. Solve the following systems of homogeneous linear equations.
-1 1 -1
=
=
x + 2 y - 2z =0 x1 + 4 x2 + 2 x3 =0 x1 - 2 x2 - x3 =0
x2
9 9
  
18 - 12 + 3 9 i) 2 x + y + 5 z =
0 ii) 2 x1 + x2 - 3 x3 = 0 x1 + x2 + 5 x3 =0
= = = 1
0  0  0 
iii)
9 9 5x + 4 y + 8z = 3 x1 + 2 x2 - 4 x3 = 2 x1 - x2 + 4 x3 =
62 63
1. Quadratic Equations eLearn.Punjab
3. Matrices and Determinants eLearn.Punjab
5. Find the value of λ for which the following systems have non-trivial solutions. Also solve
the system for the value of λ .
x+ y+z= 0  x1 + 4 x2 + λ x3 =0
 
i) 2 x + y - λ z =0 ii) 2 x1 + x2 - 3 x3 = 0
x + 2 y - 2z = 0  0 
3 x1 + λ x2 - 4 x3 =
6. Find the value of λ for which the following system does not possessa unique solution.
Also solve the system for the value of λ .
x1 + 4 x2 + λ x3 = 2 

2 x1 + x2 - 2 x3 =11 
16 
3 x1 + 2 x2 - 2 x3 =
version: 1.1
64

Matrices and Determinants: Animation 3.1: Addition of Matrix Source & Credit: Elearn - Punjab

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Matrices and Determinants: Animation 3.1: Addition of Matrix Source & Credit: Elearn - Punjab

Uploaded by

Copyright:

Available Formats

version: 1.

Animation 3.1: Addition of matrix

is called an m by n matrix (written as m % n matrix).

 b '11 b '12 b '13   b11 b21 b31 

3.1.1 Addition of Matrices

3.1.2 Scalar Multiplication

 b11a11 + b12 a21 b11a12 + b12 a22 b11a13 + b12 a23 

 2 -1 0   2 -2 3 ad - bc, that is,

3.2.1 Singular and Non-Singular Matrices

Similarly, solving (ii) and (iv), we get

Thus AA-1 = A-1A = I2

Solution : If A =   and A =  r s  , Then:

3.3 Solution of simultaneous linear equations by using matrices

Let the system of linear equations be

where a11, a12, a21, a22,+b1 and

Example 5: Solve the following systems of linear equations.

version: 1.1 version: 1.1

If A = [aij]3%3, ind the following matrices;

If A = [aij]2%3 and B = [bij]2%3, show that λ ( A + B) = λ A + λ B.

1. Commutative property w.r.t. addition: A + B = B + A

version: 1.1 version: 1.1

A-B B-A iii) (A - B) - C iv) A - (B - C)

 a11 a13  a1 j  a1n 

A = a11 - a12 + a13

or |A| = a11(a22a33 - a23a32) - a12(a21a33 - a23a31) + a13(a21a32 - a22a31)

version: 1.1 version: 1.1

Example 2: Find the cofactors A12, A22 and A32 if A =  -2 3 1  and

3.7 Properties of Determinants which Help in their Evaluation

= 1[6 - 1(-3)] + 2[(-2).2 -1.4] + 3[(-2)(-3) -12]

= ka11a22 - ka12a21 = k (a11a22 - a12a21) = k 11 12

= a11a22 - a12a21 = a11a22 - a21a12 = 11

= a12a21 - a11a22 = -(a11a22 - a12a21) (columns are interchanged)

a11 a12 a11 a12 + ka11

A=  , evaluate |A|

version: 1.1 version: 1.1

Expanding by irst column, we have

x a b+c 1 a b+c  by property 5 or 

= x b c += x1 b c+a  taking x common 

 adding the entries of C3 to the 

3.8 Adjoint and Inverse of a Square Matrix of Order n83

 a11 a12 a13   A11 A13 

If A =  a21 a22 a23  , then the matrix of co-factors of A =  A21 A23  ,  3 1 -2 

Inverse of a Square Matrix of Order n83: Let A be a non singular

2. Without expansion show that

x 1 1 1 15. Verify that (AB)t = Bt At and if

9. Find |AAt| and |AtA| if

version: 1.1 version: 1.1

From At = -A, it follows that [a’ij] = for i, j = 1, 2, 3, ......., n

which implies that a’ij = -aij for i, j = 1, 2, 3, ......., n

Alternatively we can say that a square matrix A = [aij]nxn is anti-symmetric 1 i 2  1 i 2 

hermitian or anti-hermitian if ( A ) = -A.

From ( A ) = -A, it follows that  a ij/  = [-aij]nxn

which implies that a ij/ = -aij for i, j = 1, 2, 3, ..., n. 0 0 0 0  0 0 0 1 

For diagonal elements j = i, so aii = -a ii ⇒ aii + -a ii = 0

R 0 1 -1 3  By R3 +( - 4)R2 → R '3 R 0 1 -1 3  By R1 +1.R2 → R '1

Interchanging R1 and R3 we get..

0 0 1 -1 0 0 1 -1 1 2 -2  0 0 1  1 0 6  0 0 -2 

Example 2: Find the inverse of the matrix A =  3 4 2   

Interchanging C1 and C3, we get

version: 1.1 version: 1.1

Exercise 3.4 10. Find the rank of the following matrices

is called a system of non-homogeneous linear equations in the two variables x and y if