(2 :30 PM)

(Important)
 MATRICES AND DETERMINANTS

A matrix is a rectangular array of numbers (any system of numbers), arranged in

rows and columns. If there are m rows, and n columns in a matrix, it is called an ‘m
by n’ matrix, or a matrix of order m × n. The transpose of a matrix is obtained from
A by changing its rows and columns, and columns into rows. It is denoted by A' or
AT .

Properties of matrix multiplication:

1. Non-Commutative: AB ≠ BA in general.
2. Associative: ( AB )C = A(BC )
3. Distributive: A(B C C ) = AB C AC
4. Scalar Multiplication: k = =
( AB ) (kA )B A(kB )
5. Transpose Property: ( AB )' = B ' A'

Symmetric Matrix: A matrix which is equal to its transpose, is a symmetric matrix.

=
For a symmetric matrix aij i , aij a ji for all i and j .
A =

Skew Symmetric Matrix: A matrix A such that A = -A' is called skew symmetric.
For a skew symmetric matrix A =  aij i , aij = −a ji for all i and j .

Theorem 1: A square matrix can be uniquely expressed as a sum of a symmetric

and a skew symmetric matrix.

Transformation of Matrices:
1. Row/Column Interchange: Ri ↔ R j or Ci ↔ C j . Ri , R j stand for i th and j th
rows and Ci ,C j stand for i th and j th columns.
2. Scalar Multiplication: R j → kR j or C j → kC j ; k ≠ 0 .
3. Row/Column Addition: Ri → Ri + kR j or Ci → Ci C kC j .

Inverse of a matrix: If A and B are two square matrices of the same order such
that AB
= BA = I , the unit matrix of the same order as A or B , then B is called the
multiplicative inverse of or simply the inverse of A , written as A−1 . A is also called
inverse of B , written as B −1 .

Theorem 2: Inverse of a matrix, if it exists, is unique.

Theorem 3: If A and B are invertible matrices of the same order, then

( AB )−1 = B −1 A−1
Determinant of a square matrix:
a b c
d e f = aei − afh + bfg − bdi + cdh − ceg is the determinant of the square matrix
g h i
 a b ch
 
d e f  .
e g h i i

Properties of Determinants:
1. Swapping Rows/Columns: Changes sign of determinant
2. Identical Rows/Columns: det( A) = 0
3. Row/Column Scaling: det(kA) = k n det( A) (for n × n matrix)
4. Row Addition: No change in determinant
5. Triangular Matrix: det( A) = Product of diagonal elements
6. Factor Property: If the determinant vanishes for x = a , then ( x − a) is a
factor of det( A) .

Singular matrix: A square matrix A is called singular if det A = 0 . If det A ≠ 0 the

matrix A is called a non-singular matrix or a regular matrix.

Adjoint of a matrix: If A =  aij i is a square matrix, then the transpose of the

n× n

matrix  Aij i of which the elements are cofactors of the corresponding element
n×n

in | A | , is called the adjoint of A . It is denoted by adjA .

Thus, adjA =  Aij i
n×n

 a11 a12 a13   A11 A21 A31 

   
If A =  a21 a22 a23  , then adjA =  A12 A22 A32  where
 a31 a32 a33   A13 A23 A33 
Aij is the cofactor of aij in | A | .

Theorem 4: If A is a square matrix, then A(adjA

= =
) |A | I (adjA)A .

Theorem 5: Let A be a square matrix. Then A−1 exists if A is non-singular and the
inverse is given by
1
A−1 = adjA
| A|
System of linear equations and solution:
Suppose we have the following system of equations:
a1 x ++
b1 y c1 z =
d1
a2 x ++
b2 y c2 z =d2
a3 x ++
b3 y c3 z =d3
  a1 b1 c1  x  d1 
     
We write A =  a2 b2 c2  , X =  y  , B = d2 
 a3 b3 c3   z  d3 

In the matrix form becomes AX = B .

Theorem 6: The system of equations has a unique solution if and only if the matrix
is non-singular, and in this case, the matrix solution of the equations is given by
1
= X adjA ⋅ B
| A|

Cayley-Hamilton Theorem: Every square matrix satisfies its characteristic

equation.
c1 An−1 c2 An−2 .... cn I =
An ++++ 0

Eigenvalues and Eigenvectors:

Eigenvalue Equation: Ax = λ x
Characteristic Equation: | A − λI |=
0
Diagonalization: If A has n linearly independent eigenvectors, it can be written as
A = PDP −1 where P contains eigenvectors and D is a diagonal matrix of
eigenvalues.

Rank of a Matrix:
Rank: Maximum number of linearly independent rows or columns.
Full Rank: If rank ( A) = min(m, n) , A has full rank.

Rank-Nullity Theorem: If A is a matrix of order m × n , then

Rank of A + Nullity of A = Number of columns in A = n

LU Decomposition: A square matrix A can be decomposed as LU , where L is the

lower triangular matrix and U is the upper triangular matrix.

Example 1: Let ω be a complex cube root of unity with ω ≠ 1 and P = [ pij ] be a n × n

matrix with pij = ω
i+ j
. Then P 2 = 0 , find the least prime number such that n should
not be a multiple of that prime number.

Solution:
P = [ pij ]
pij = ω i + j
where ω is a complex cube root of unity such that ω ≠ 1 , i.e., ω satisfies:
ω 3 = 1,1 ++
ω ω2 = 0
 n
(P 2 )ij = ∑ Pik Pkj
k =1

Substitute Pik = ω i +k
and Pkj = ω k+ j
,
n
(P 2 )ij = ∑ ω i ++
ω
k k j

k =1
n
(P 2 )ij = ∑ ω ( i ++
j ) 2k

k =1
n
(P 2 )ij = ω i + j ∑ ω 2 k
k =1
n
S = ∑ω 2 k
k =1

n
The sum S = ∑ ω 2 k is a geometric series with common ratio ω 2 and n terms. The
k =1

sum of a geometric series is:

ω 2 (1 − (ω 2 )n )
S=
1−ω2
For P 2 = 0 , we require that this sum equals zero, i.e.,
ω 2 (1 − (ω 2 )n )
=0
1−ω2
which holds if and only if (ω 2 )n = 1 , or equivalently.
ω 2n = 1
Since ω 3 = 1 , it follows that ω 2 n = 1 if and only if 2n is a multiple of 3, i.e.,
2 n ≡ 0 (mod3)
or equivalently,
n ≡ 0 (mod3)
This means that for P 2 = 0 , n must be a multiple of 3.

Example 2: How many 3 × 3 matrices M with entries from {0, 1, 2} are there, for
which the sum of the diagonal entries of MTM is 4?

Solution:
a b ch
 
M = d e f
e g h i i
a d gh
T  
M = b e h
e c f i i

Then,
  a d gh  a b ch
T   
M M =  b e h  d e f 
e c f i i e g h i i
 a2 ++++++
d 2 g 2 ab de gh ac df gi h
 
MT M =  ba ++++++
ed hg b2 e2 h2 bc ef hi 
 ca ++++++
fd ig cb fe ih c 2 f 2 i 2 i
e
Let S be the sum of diagonal entries.
S = a2 ++++++++
d 2 g 2 b 2 e 2 h2 c 2 f 2 i 2
⇒ S = a2 ++++++++
b 2 c 2 d 2 e 2 f 2 g 2 h2 i 2
Now, we will have to find the combinations of S = 4 by putting the values of
a, b, c, d , e,f , g, h and i as 0,1 and 2 .
First, 4 =12 ++++++++12 12 12 02 02 02 02 02
Second, 4 = 2 2 ++++++++ 02 02 02 02 02 02 02 02
In both the above cases, we have 9 entries.
In this first case, we have 4 elements of 1 and 5 elements of 0.
Therefore, the total number of combinations in the first case:
9 9! 9× 8× 7 × 6
= C4 = = 126
4!(9 − 4)! 4 × 3 × 2 × 1
In the second case, we have one element of 2 and 8 elements of 0.
Therefore, the total number of combinations in the second case:
9 9! 9
C=1 = = 9
1!(9 − 1)! 1
Hence, the total number of combinations is 126 + 9 =
135 .

Example 3: Let P be a matrix of order 3 × 3 such that all the entries in P are from
the set {-1, 0, 1}. Then, find the minimum possible value of the determinant of P .

Solution:
 a b ch
 
P = d e f 
e g h i i
Therefore,
e f d f d e
| P |= a × − b× + c×
h i g i g h
The maximum possibility of P can be 6 if we have the following:
e f
= ±2
h i
d f
= ±2
g i
 d e
= ±2
g h
e f
However, if = ±2
h i
d e
and = ±2 , then
g h
d f
=0
g i
Therefore, | P |≠ −6 .
Thus, the next possible value of | P | is -4.

Example 4: Given − x = cy + bz , −y = az + cx , − z= bx + ay where x, y , z , are not all

b2 c 2 − 2 abc .
zero. Find the value of a2 ++

Solution:
Rewriting,
x ++
cy bz =
0
cx ++
y az =
0
bx ++
ay z =
0

Using the condition for concurrency

1 c b
c 1 a =0
b a 1
1(1 − a2 ) − c(c − ab) + b(ac − b) =
0
1 − a2 − c 2 ++
abc abc − b2 =
0
a2 ++
b2 c 2 − 2 abc =
1

Example 5: Find the total number of distinct x ∈ R for which

x x2 1+ x3
2x 4x2 1 + 8x 3 =
14 .
3 x 9x 2 1 + 27 x 3

Solution:
x x2 1+ x3
2x 4x2 1 + 8x 3 =
14
3x 9x 2 1 + 27 x 3
 1 1 1+ x3
⇒ x3 2 4 1 + 8x 3 =
14
3 9 1 + 27 x 3
R2 → R2 − 2R1 , R3 → R3 − 3R1 , we get

1 1 1+ x3
⇒ x3 0 2 6x3 − 1 =
14
0 6 24 x 3 − 2
⇒ x 3 (2(24 x 3 − 2) − 6(6 x 3 − 1)) =
14
⇒ x 3 (12 x 3 + 2) =
14
⇒ 6x6 + x3 =
7

Put x 3 = t ,
⇒ 6t 2 + t − 7 = 0
−7
⇒ t = or t = 1
6
−7
⇒ x 3 = or x = 1
3

6
−7 1
( )3 or x = 1
⇒x=
6
So, x has 2 distinct values.

 2k − 1 2 k 2 k 
 
Example 6: Let k be a positive real number, = and let A  2 k 1 −2 k  and
 
 −2 k 2 k −1 

 0 2k − 1 k 
 
B= 1 − 2 k 0 2 k  . If det(adjA) + det(adjB ) =
729 , then find [k ] .(Note: adjM
 
 − k −2 k 0 

denotes the adjoint of a square matrix M and [k ] denotes the largest integer less
than or equal to k )

Solution:
0 (B is a skew-symmetric matrix of order 3)
(2 k + 1)3 ,| B |=
| A |=
Let det(adjA) =| A |n−1
((2 k + 1)3 )2 =
729
((2 k + 1)3 )2 =
36
2k + 1 = 3
k =1
[k ] = 1
 EXERCISE

1 + sin2 x cos2 x 4sin2 x

1.=
If f ( x ) sin2 x 1 + cos2 x 4sin2 x , what is the least value of f ( x ) ?
sin2 x cos2 x 1 + 4sin2 x
 1 0 0
 
2. Let P = 3 1 0 and I be the identity matrix of order 3. If Q = qij i is a matrix
9 3 1 
q + q32
such that P 100 − Q =I , then calculate 31 .
q21

3. Let A be a 3 × 3 real matrix such that A3 − 6 A2 + 11A − 6I =

0 and det( A) = 6 . Let
−1
=
B A + 2I . Find det(B ) .

SOLUTIONS

1.
1 C sin2 x cos2 x 4sin2 x
2 2
=f (x) sin x 1 C cos x 4sin2 x
2 2
sin x cos x 1 C 4sin2 x
C1 → C1 C C2
2 cos2 x 4sin2 x
( x ) 2 1 C cos2 x
f= 4sin2 x
1 cos2 x 1 C 4sin2 x
R2 → R2 − R1 , R3 → R3 − R1
2 cos2 x 4sin2 x
f (x) = 0 1 0
−1 0 1
f ( x )= 2 C 4sin2 x

Thus, the least value of f ( x ) is f ( x ) =2−4 =−2 . ( −1 ≤ sin2 x ≤ 1)

 1
0 0
 
2. P = 31 0
9
3 1 
1
0 0  1 0 0
  
P 2 = 3
1 0  3 1 0 
9
3 1  9 3 1 
 1 0 0
2  
=
P  3+3 1 0
32 +++
32 9 3 3 1 

 1 0 0  1 0 0
 3  
P =  6 1 0  3 1 0 
27 6 1  9 3 1 
 1 0 0
3  
=
P  6+3 1 0
27 +++18 9 6 3 1 
 
 1 0 0
 
∴P n =  3n 1 0
 n(n + 1) 
 9 3n 1 
 2 
100
Now,= Q P −I
 1 0 0  1 0 0
   
=Q  300 1 0  − 0 1 0 
45450 300 1  0 0 1 
 0 0 0
 
Q =  300 0 0
45450 300 0
q31 + q32 45450 + 300
∴ = = 152.5
q21 300
q +q
The value obtained for 31 32 is 152.5
q21
3. A3 − 6 A2 + 11A − 6I =
0
This is reminiscent of a minimal polynomial or characteristic polynomial.
Let’s factorise:
x 3 − 6 x 2 + 11 x − 6 = ( x − 1)( x − 2)( x − 3)
So, this suggests the eigenvalues of A are:
λ1 = 1
λ2 = 2
λ3 = 3
Since the minimal polynomial has a distinct linear factor and is satisfied by
A , we conclude: A is diagonalizable

Eigenvalues of A : 1,2,3
B A−1 + 2I
=
Since eigenvalues of A are λ1 = 1 , λ2 = 2 , λ3 = 3 , the eigenvalues of A−1 are
1 1 1
, ,
1 2 3
µ1 = 1
1
µ2 =
2
1
µ3 =
3

Then eigenvalues of =
B A−1 + 2I are:
µ1 + 2 =
3
5
µ2 + 2 =
2
7
µ3 + 2 =
3

So,
3
5 7
det(B ) = ∏ ( µi + 2) = 3 × ×
i =1 2 3
35
det(B ) =
2

35
The value of det(B ) is .
2