Lecture # 4 Linear Algebra And Analytical Geometry

SPECIAL SQUARE MATRICES

1. Periodic Matrix
A square matrix A is said to be periodic if A k +1 =A⋅¿ ¿
k+1
If k is the least positive integer for which A =A , then k is called the “period” of A.

EXAMPLE: Show that the matrix

1 −2 −6

is a periodic matrix having period 2.

[
A= −3 2 9
2 0 −3 ]
Solution:Consider

1 −2 −6 1 −2 −6 1+6−12 −2−4−0 −6−18+18

2
[
A = −3 2
2
9 −3 2
0 −3 2 0 −3
9
][ ] [ = −3−6+18 6+4+0 18+18−27
2−0−6 −4+0−0 −12+0+9 ]
−5 −6 −6

[
= 9 10 9
−4 −4 −3
Now consider,
] ≠A.

−5 −6 −6 1 −2 −6
A 3 =A 2 A= 9
[ 10 9 −3 2
−4 −4 −3 2
9
0 −3 ][ ]
−5+18−12 10−12−0 30−54+18 1 −2 −6
[
= 9−30+18 −18+20+0 −54+90−27
−4+12−6 8−8−0 24−36+9 ] [
= −3 2 9 =A⋅¿ ¿
2 0 −3 ]
We observe that A 3=A , hence A is a periodic matrix having period 2.

2. Idempotent Matrix
2
A square matrix A is said to be Idempotent if A = A⋅¿ ¿
EXAMPLE: Show that
2 −2 −4

Solution: Consider
[
A= −1 3 4
1 −2 −3 ] is an Idempotent matrix.

2 −2 −4 2 −2 −4 4+2−4 −4−6+8 −8−8+12

2
[
A = −1 3 4 −1 3 4
1 −2 −3 1 −2 −3 ][ ] [ = −2−3+4 2+9−8 4+12−12
2+2−3 −2−6+6 −4−8+9 ]
 2 −2 −4

[
= −1 3 4
1 −2 −3 ]=A
Since A2 = A, the given matrix A is an Idempotent Matrix.
NOTE:Identity matrix is always a periodic of period one. It is also an Idempotent matrix for I2 = I.
It may also be noted that an idempotent matrix is periodic matrix with period 1.

3. Nilpotent Matrix
k
A square matrix A is said to be nilpotent if A =O, k ∈N⋅¿ ¿
k
If k is the least member of N such that A =O , then k is called the “index” of A.

ab b 2
EXAMPLE: Show that
A= [ −a2 −ab ] is a nilpotent matrix of index 2.
Solution: Consider
ab b2 ab b2
A2=
[ −a2 −ab −a 2 −ab ][ ] ¿
ab3 −ab3
a2 b2−a2 b2
= 3 [3 2 2
−a b+a b −a b +a b2 2 =
0 0
0 0
=O⋅¿
][ ]
Since, A2 = O, the matrixA is a nilpotent matrix of index 2. You may observe that | A | = 0.

[ ]
−3 −4
−1 3 4
EXAMPLE:Show that 1 −3 −4 is nilpotent.
1 −3 −4

Solution: Let
[
A= −1 3 4
1 −3 −4 , then ]
1 −3 −4 1 −3 −4

[
A 2 =A⋅A= −1 3 4 −1 3 4
1 −3 −4 1 −3 −4 ][ ]
1+3−4 −3−9+12 −4−12+16 0 0 0

[
= −1−3+4 3+9−12 4+12−16 = 0 0 0 =O
1+3−4 −3−9+12 −4−12+16 0 0 0 ][ ]
Thus, the given matrix is nilpotent of index 2. You may observe that | A | = 0.
4. Involutory Matrix
A square matrix A is said to be Involutory matrix if A 2 =I⋅¿ ¿
0 1 −1

EXAMPLE: Show that

Solution: Consider
[
A= 4 −3 4
3 −3 4 ] is Involutory.
 0 1 −1 0 1 −1 0+4−3 0−3+3 0+4−4 1 0 0
2
[ ][ ][
A = 4 −3 4 4 −3 4 = 0−12+12 4+9−12 −4−12+16 = 0 1 0 =I⋅¿ ¿
3 −3 4 3 −3 4 0−12+12 3+9−12 −3−12+16 0 0 1 ][ ]
Since, A2 = I, henceA is anInvolutory matrix.
NOTE: Since A2 = IA A = I. Pre-multiplying both sides by A-1, we get
A-1(A.A) = A-1 . I  (A-1A) .A = A-1 I. A = A-1 A = A-1.
This shows that if a matrix A is involutory matrix then it is equal to its inverse.
SPECIAL MATRICES IN GENERAL
1. Transpose of a Matrix
Let A be any matrix of order m × n. The matrix obtained by interchanging rows and columns of Ais called
thetranspose of a matrix. The transpose of matrix is denoted by At. For example, if
1 0
A=
[ 1 2 3
0 5 7 ]2 ×3
then
,

Thus by definition, if matrix A = [ai j], then At = [aji]

t
A=2 5
3 7 [ ] 3×2
¿

Properties of Transpose of Matrices

If the matrices A and B are conformable for the sum A+B and the product AB, then
t t t t t t
(i) ( A+B )t =A +B (ii) A t = A (iii) ( kA )t =kA , where k ∈R (iv)
( ) ( AB )t =B A ⋅¿ ¿
Proof: (i) LetA andB be m × n matrices. Then by definition A + B= [ai j + bi j] be m × n matrix, hence (A + B)t will be n
× m matrix. Now
(i, j) th element of (A + B)t = (j, i) th element of (A + B) = aji + bji. And,
(i, j) th element of At + Bt = (i, j) th element of At + (i, j) th element of Bt
= (j, i) th element of A+ (j, i) th element of B = aji + bji
Thus (i, j) th element of (A + B)t= (i, j) th element of At + Bt
Hence, (A + B)t = At + Bt.
(ii) (i, j) th element of ((A)t)t= (j, i) th element of At
= (i, j) th element of A.
Hence, ((A)t)t = A.
(iii) (i, j) th element of (kA)t = (j, i) th element of kA
= k ajI
t
= (i, j) th element of kA .
Hence, (kA)t = k At.
(iv) LetA = [ai j] be an m ×n matrix and B = [bi j] be an n × p matrix.
Then A B is of order m × p and so (AB)tis of order p × m.
Now Btis of order p ×n andAt is of order n × m. Hence Bt Atis of order p ×m.
n
∑ a jk b ki
(i, j) th element of (AB)t = (j, i)th element of (AB) = k =1
(i, j) th element of BtAt= sum of the products of the corresponding elements of ith row of Bt and jth column
ofAt.
= sum of the products of the corresponding elements of ith column of B and jth row
ofA.
 n
∑ a jk b ki
= k =1

t t t
Hence ( AB ) =B A ⋅¿ ¿
Corollary: (A1A2 …An)t = Ant. … A2tA1t
2. Conjugate of a Matrix
A complex matrix obtained by replacing its elements by their corresponding complex conjugates is called the

conjugate of A and is denoted by A⋅¿¿ For example, if

2+3i 4 5i 2−3i 4

[ ] [ ]
−5i
A= 0 4i 8 , A= 0 −4i 8
4−3i 0 7+3i then
4+3i 0 7−3i is the conjugate matrix of A.
3. Tranjugate or Transpose of a Conjugate Matrix
The transpose of the conjugate of a matrix A is called the tranjugate or transposed conjugate of A. It is denoted by
( A ) t⋅¿ ¿ For example, if
3+4i 5−6i 2+3i 3−4i 5+6i 2−3i

6[
A= 4−5i 7 8i ,
5+6i 2−3i ] then
A = 4+5i
6 [ 7
5−6 i
−8i
2+3i ]
3−4i 4+5i 6

and
[
( A ) t = 5+6i 7 5−6i ⋅¿ ¿
2−3i −8i 2+3i ]
Properties of Tranjugate Matrix: If A and B are two complex matrices confirmable for addition and
multiplication, then

(i) A±B= A ±B (ii)

t
( A ) =( A
t
) (iii) A⋅B = A⋅B (iv) kA =k A , where k is a scalar, real or complex.
_______
__

(v) (( A) ) =A (vi) A +A=Re al matrix (vii) A−A=Purely imaginary matrix

Proof: (i) Let A = [aij + ibij] and B = [cij + idij]. Then A = [aij - ibij] and B = [cij - idij]. Now
A + B = [aij + ibij]+[cij + idij]= [aij + cij] +i [bij + dij]

 A +B =[aij + cij] -i [bij + dij] = [aij - ibij] + [cij - idij] = A +B

Proofs of parts (ii) to (vii) are left as an exercises.
4. Symmetric Matrix
t
Any square matrix A is said to be symmetric if A =A⋅¿ ¿ For example, if
2 3 2 3 =A⋅¿ ¿
A=
[ ]
3 5
,
then
[ ]
At =
3 5
5. Skew-Symmetric Matrix
t
Any square matrix A is said to be skew-symmetric if A =−A⋅¿ ¿ For example, if
 0 1 2 0 −1 −2 0 1 2

6. Hermitian Matrix
[
A= −1 0 −3 ,
−2 3 0 ] then
At= 1
[ 0 3
2 −3 0
=−
][
−1 0 −3 =−A⋅¿ ¿
−2 3 0 ]
__
t
A square matrix A for which ( A ) =A is called a Hermitian matrix. For example, if
a b−ic a b +ic
A=
[ b+ic d
,
] then
A=
[ b−ic d ]
and
( A )t =
[b+ica b−ic =A⋅¿ ¿
d ] So, A is a Hermitian matrix.
7. Skew-Hermitian Matrix
__
t
A square matrix A for which ( A ) =−A is called a skew-Hermitian matrix. For example, if
i 1−i 2 1+i 2

[ ] [ ]
−i
A= −1−i 3i i , A = −1+i −3i −i
−2 i 0 then
−2 −i 0

−i −1+i −2 i 1−i 2
t
[
( A ) = 1+i −3i −i =− −1−i 3i i =− A⋅¿ ¿
2 −i 0 −2 i 0 ][ ]
1 1+i 2+3i i 1+i 2−3i

EXAMPLE 01: If
[
A= 1−i
2−3i i
2 −i
0 ] and
[
B= −1+i 2i
−2−3i −1
1 ,
0 ] then show that

(i) iB is Hermitian(ii) A is Hermitian (iii) B is Skew – Hermitian.

Solution:(i)
i 1+i 2−3i −1 i−1 2i+3

[
B= −1+i 2i
−2−3 i −1
1 ⇒iB= −i−1 −2
0 −2i+3 −i
i ⋅¿ ¿
0 ] [ ]
If iB is Hermitian, then it should satisfy the condition: ( iBt ) =iB⋅¿¿ Now,
__ −1 −i−1 −2i+3 −1 i−1 2i+3
iB = i−1 −2
2i+3 i [ −i = −i−1 −2
0 −2i+3 −i
i
0 ][ ]
−1 i−1 2i+3 −1 i−1 2i+3

[ ][ ]
__
t
∴(iB ) = −i−1 −2 i = −i−1 −2 i =iB⋅¿ ¿
−2i+3 −i 0 −2i+3 −i 0
Hence iB is Hermitian.

(ii) We have
1 1+i 2+3i 1 1−i 2−3i

[
A= 1−i 2
2−3i i
−i
0
⇒ A = 1+i 2
2+3i −i 0
i ⋅¿ ¿
] [ ]
Again, if A is Hermitian then it should verify the following condition: ( A )t ) =A⋅¿ ¿ Now,
______
1 1+i 2+3i 1 1−i 2−3i 1 1−i 2−3 i __

[ ][ ][ ]
__
t
( A) = 1−i 2 −i = 1+i 2 i ⇒ 1+i 2 i =A⋅¿ ¿
2−3i i 0 2+3i −i 0 2+3 i −i 0
Hence A is Hermitian.
(iii) We have
i 1+i 2−3i −i 1−i 2+3i

[
B= −1+i 2i
−2−3 i −1
1 ⇒B= −1−i −2i 1 ⋅¿ ¿
0 −2+3i −1 0 ] [ ]
t
If B is Skew – Hermitian then it should verify the following condition: −( B) )= B⋅¿ ¿ Now,
___________
−2+3 i 1−i 2+3i

[ ][ ]
__ −i −1−i −i __
t
− (( B ) ) = − 1−i −2i −1 = −1−i −2i 1 =B
2+3i 1 0 −2+3i −1 0

Hence, B is skew-Hermitian.
8. Orthogonal Matrix
A square matrix A is said to be orthogonal if AtA = I = A At
EXAMPLE 02: Show that the matrix
cosθ 0 sinθ

is orthogonal.
[ 0 1 0
sinθ 0 cosθ ]
Solution: By definition consider,
cos θ 0 -sinθ cosθ 0 sinθ
t
AA = 0
[
1 0 0 1 0
sinθ 0 cosθ -sinθ 0 cosθ
2 2
][ ]
cos θ+sin θ 0 cosθ sin θ−cos θ sin θ 1 0 0
=0
[ 1 0
2 2
cosθ sin θ−cos θ sin θ 0 cos θ+sin θ
= 0 1 0 =I
0 0 1
Similarly we can show that At A = I. Hence given matrix is orthogonal.
][ ]
NOTE: Since At A = IAt = A-1

−1 2 2

EXAMPLE 03: Prove that

Solution:
1
A= 2 −1 2
3 [
2 2 −1 ] is orthogonal.

−1 2 2 −1 2 2
A=
1
3[ 2 −1 2
2 2 −1
⇒ A
t 1
=
3 ] [
2 −1 2 ⋅¿ ¿
2 2 −1 ]
Then
−1 2 2 −1 2 2 9 0 0 1 0 0
A t A=
[
1
9
t
2 −1 2 2 −1 2
2 2 −1 2 2 −1][
=
1
9
0 9 0
0 0 9
=
9
9 ][ ][ ]
0 1 0 =I⋅¿ ¿
0 0 1
Since, A A=I , therefore matrix A is orthogonal.
Theorem: If A andB are orthogonal matrices, then AB and BA are also orthogonal matrices.
Proof: Since A and B are orthogonal matrices, we have,AAt= I and B Bt= I. We have to prove that AB and BA are
also orthogonal.
Consider, (AB)(AB)t = AB. BtAt= A (B Bt)At= A(I) At= A At = I  AB is orthogonal.
Similarly, consider (BA)(BA)t = BA. At Bt= B (BA At)Bt = B(I) Bt = B Bt = I  BA is orthogonal.
9. Unitary Matrix

A square matrix A is said to be unitary if ( At ) A=I⋅¿ ¿

Theorem: If A andB are n – rowed unitary matrices, then AB and BA are also unitary matrices.
Proof: Exercise
1+i −1+i

EXAMPLE 04:Show that

Solution:
[ ]
A= 2
1+i
2
2
1−i
2 is a unitary matrix.

1+i −1+i 1+i 1+i 1−i 1−i

A= 2

Now
1+i
2
[ ] [2
1−i
2
⇒A = 2
t

−1 +i
2
1−i
2
] [
2 ⋅¿ ¿

Also
( A )= 2
t

−1−i
2
2 ¿
1+i
2
]
1−i 1−i 1+i −1+i
(A )A= 2
t

−1−i
2
[ ][ 2
1+i
2
2
1+i
2
2
1−i
2
] ( )( )[
=
1
2
1 1−i
2 − ( 1+i )
1−i 1+i −1+i
][
1+i 1+i 1−i

( 1−i 2 ) + ( 1−i2 ) −( 1−i )2 + ( 1−i )2 1 4 0 4 1 0 1 0

]
( At ) A = 1
[ ]
4 −( 1+i )+ (1+i )2 ( 1−i2 ) + ( 1−i 2 )
= [ ] [ ][ ]
=
4 0 4 4 0 1
=
0 1
=I ¿

Hence A is a unitary matrix.

1. 6 SOME IMPORTANT RESULTS FOR SQUARE MATRICES
1. If A is a square matrix, prove that A + Atis a symmetric matrix and A – At is a skew-symmetric matrix.

Proof: Let C = A + At then C t=(A + At ) t = At + (At ) t = At + A = A + A t = C. Since C t = C,hence C is symmetric. But C = A

+ At, hence A + At is symmetric.

Now let D = A - Atthen D t=(A - At ) t = At - (At ) t = At - A = -(A - A t ) = - D. Since D t= -D,hence D is skew-symmetric. But
D = A - At, hence A - At is skew-symmetric.

2. If A be a skew-symmetric matrix, then its diagonal elements are zero.

Proof: Given that A is a skew-symmetric matrix, hence At = - A. This implies thataij = - aijfor all i, j. This must be true
for the diagonal elements as well. Thus, aii = - aii oraii + aii = 0 or 2 aii = 0 aii = 0. This proves that for a skew-
symmetric matrix A, its diagonal elements are zero.
EXAMPLE 01: Let
2 8 5 2 1 7 4 9 12 0 7 -2

[ ] [ ]
7 9 5
t

5 4 5
t

12 13 10
t
A= 1 3 4 ⇒ A = 8 3 9 ∴ A + A = 9 6 13 and A− A = -7 0 -5
2 5 0 [ ] [ ]
We see that A + Atis a symmetric matrix and A – Atis skew-symmetric matrix. We also see that all the diagonal
elements of a skew-symmetric matrix are zero.
t t
3. If A is a square complex matrix, then A +( A ) is Hermitian and A−( A ) is skew-Hermitian matrix.
Proof: Let
________ ___ __ ____
¿t t __
t
−
( )
C=A+ ( A ) ⇒C = A + ( A ) =A + ( A ) = A +( A ) =A +A t
t
¿ ¿ ¿ ¿
[∵(A )=A ]
−t ¿ t −t
tt
() t
⇒ C =( A +A ) =( A ) + (( A) ) =( A ) +A=C. Since C =C, hence C is a Hermitian matrix.
t t ()
Now let,
________ ___ __
¿t ¿ ¿t ¿
−
( )
D=A− ( A )t ⇒D = A − ( A ) =A − ( A ) = A −( A ) =A −A t
t
¿

− t ¿ t
tt
(⇒ D) =( A −A ) =( A) − (( A) ) =( A )t−A=−( A−( A) )=−D.
t t t

Since,
t
( D ) =−D , hence D is a skew-Hermitian matrix.
4. If A be a Hermitian matrix, its diagonal elements are real. If A be a skew-Hermitian matrix, then its diagonal
elements are either zero or purely imaginary.
t
Proof: Let A be a Hermitianmatrix  ( A ) =A . Now let its diagonal element be aii = α + i β. This
aii aii
implies that = α - i β. But aii =  α + i β = α - i β  2 i β = 0  β = 0. Thus aii = a which is a real
number.
t
Let A be a skew-Hermitianmatrix  ( A ) =−A . Now let its diagonal element be aii = α + i β. This implies that aij
= α - i β. But aij = - aii

 α + i β = - ( α - i β )  α + i β = - α + i β 2 α = 0 α = 0.

Thus diagonal elements of Skew-Hermitian matrix are purely imaginary, that is, aii = i β. If
β = 0, then diagonal elements are zero.
05: If A and B be any two symmetric matrices then (AB) t = A B if and only if they commute.
Proof: Given that A and B are symmetric matrices, that is, At = A and B t = B.
Now let us assume that A and B commute, that is;
 AB=BA (1)

Then by definition, (A B)t = Bt At = B A [ Because At = A and B t = B]

= A B [By equation (1)]

Thus, (A B) t= A B. This proves that AB is symmetric.

Now let us assume that (AB) t = A B.

 B t At = A B
 BA =AB [Because At = A and Bt = B]

This shows that A and B commute.

 i 2-3i 4+5i 
A = 6+i 0 4-5i 
 -i 2+i 
EXAMPLE 02: Express  2-i
as the sum of Hermitian and skew – Hermitian matrices.
Solution: We have
 1 i 2 5  5i   1 i 2 5  5i 
A  2i 2  i 4  2i  , A   2i 2  i 4  2i 
1  i 4  1  i 4 
 7 
  7 

 1  i 2i 1  i 
 A  t   2 2  i 4 
5  5i 4  2i 
 7 
Thus,
 1 i 2 5  5i   1  i 2i 1  i   2 22i 46i 
t  2i 2  i 4  2i    2  
A   A  2i  4    2  2i 4 2i 
1  i 4   
7   4 6i 2i
 7  5  5i 4  2i 14 
 1 1i 23i 
1
2 t 

A A   1i

2 i 


 23i i 7 
 (i)
Similarly,
1+i 2 5−5 i 1−i −2 i −1−i 2i 2+2i 6−4 i
A−( A ) =
t

[ 2 i 2+i 4 +2i − 2
−1+i −4 7
2−i
5+5 i 4−2 i][
−4
7 ] [ = −2+2i 2i
−6−4 i −8+2i
8+2i
0 ]
i 1+i 3−2i


1
2
{ A−( A ) t } =
[ −1+i i
−3−2i −4+i
4+i
0 ] (ii)
Adding (i)and(ii), we have
1 1−i 2−3i i 1+i 3−2 i
1
2

Hermitian Matrix
{ A+ ( A ) t }+ 1
2
t

[
{ A−( A ) }= A= 1+i 2
2+3i −i
Skew – Hermitian Matrix
7
i + −1+i i
][
−3−2 i −4 +i
4+i ⋅¿ ¿
0 ]
1.7 PARTITIONING OF MATRICES
Definition: A matrix obtained by deleting some of the rows and columns of a matrix A is said to be
sub- matrix of matrix A.
4 1 0

For example, if
[ ]
A= 5 2 1 ,
6 3 4 then
4 1
[ ] ,[ ]
5 2
5 2
6 3 and
1 0
[ ]
2 1 are sub-matrices of A.
Definition: Partitioning
A matrix may be subdivided (partitioned) into sub matrices in many different ways and the manner in which it is to
be partitioned is often indicated by a system of horizontal and vertical dotted lines. Calculation of matrices can be
greatly simplified if given matrices are represented as blocks of sub matrices, to be treated as elements of the
given matrices. The order of sub matrices is to be so chosen that the addition, subtraction and products between
them must be confirmable. For example, let

2 1 0 4 1

[
A= 1 0 2 3 4
4 5 1 6 5 ]
It may be5 partitioned as under:
2 1 ⋮ 0 4 1
A=
[1
⋯
4
0
⋯
5
⋮
⋮
⋮
2
⋯
1
3
⋯
6
4
][
⋯ A
5
A
= 11
21
A12
A22
,
] where

A 11=
[21 10 ] is a 2×2 sub matrix,
A 12=
[02 4 1
3 4 ] is a 2×3 sub matrix,

A 21=[ 4 5 ] is a 1×2 sub matrix,

A 22=[ 1 6 5 ] is a 1×3 sub matrix.
Definition: Let A and B be two m×n matrices. A andB are said to be identically partitioned if the resulting sub

matrices
A ij and
B ij of the two matrices are of the same order. For example,
−1 4 ⋮ 3 a b ⋮ c
A=
[
⋯ ⋯ ⋮ ⋯ A 11
6 −1 ⋮ 0
−7 8 ⋮ 2
=
A 21

are identically partitioned. Here

][ A12
A 22 ] and
B=
⋯
d
g
[ ][ ⋯
e
h
⋮
⋮
⋮
⋯ B11
f
i
=
B21
B 12
B 22 ]
A 11 and
B 11 are 1×2 sub matrices,
A 12 and
B 12 are 1×1 sub matrices,
A 21 and
B 21 are 2×2 sub matrices,
A 22 and
B 22 are 2×1 sub matrices.

Moreover,
A 11+B 11 , A 12 +B12 , A 21 +B21 , and
A 22 +B22 are all defined. We have
−1+a 4+b ⋮ 3+c
A +B=

EXAMPLE 01: Let

A 11+B 11 A12 +B12
[
A 21+B 21 A 22 +B22
=
⋯ ⋯ ⋮ ⋯
6+d −1+e ⋮ 0+f
−7+g 8+h ⋮ 2+i
]
⋅¿ ¿
[ ]
 1 2

A=
1 0 0

[
0 1 0
2 −3 1
2 −4 3
0
0
3
7
0
0
5
6
and
]
partitioned blocks for A and B to calculate AB.
[ ]
3 −4
B= 0 5
6 1
−2 1
be matrices of order 4 ×5 and 5 × 2 respectively. Use suitable

Solution: We partition the given matrices in the following way:

1 2

[
1 0 0 ⋮ 0
0 1 0 ⋮ 0
A= ⋯ ⋯ ⋯ ⋮ ⋯
2 −3 1 ⋮ 3
2 −4 3 ⋮ 7
whereJ, O, L, M, P and Q are of suitable orders.
0
0
⋯=
5
6
]J O
L M[ ]
and
B=

[ ]
3 −4
0
⋯ ⋯
6
−2 1
5

1
P
= ,
Q []

Then

AB=
[ JL MO ][QP ]=[LPJP+OQ
+MQ ] (i)
Now
1 2
JP= [ 1 0 0
0 1 0
3 −4 =
0 5
][
1 2
3 −4
,
][ ] OQ=
0 0 6 1 0 0
[ ][ ] [ ]
=
0 0 −2 1 0 0
,

1 2
LP=
2 −3 1
[
2 −4 3
3 −4 =
0 5
−7 21
−10 35
, ][ ] [ ] and
MQ=
[37 56 ][−26 11 ]=[308 138 ]⋅¿ ¿
Therefore,

JP+OQ=
[13 −42 ]+[00 00 ]=[13 −42 ] and
LP+MQ=
[−10−7 2135 ]+[308 138 ]=[201 2948 ]⋅¿ ¿
Hence from(i), we have
1 2

1 −1 0
AB=
3
1
20
[ ] −4
29
48
⋅¿ ¿

EXAMPLE 02: Let

[
A= −1 1 0 ⋅¿ ¿
1 −1 −2 ] Compute A8 using block multiplication.
Solution: We partition the given matrix in the following way:
1 −1 ⋮ 0
A=
[−1 1 ⋮ 0
⋯ ⋯ ⋮ ⋯ Y Z
1 −1 ⋮ −2
=
X O
,

where
X= ][
1 −1
−1 1
0 ]
,O= ,Y =[ 1 −1 ]
0 [
and Z= [−2]⋅¿ ¿
] []
 X O X O X2 O
Then
A=
2
[ ][ ] [
=
Y Z Y Z YX+ZY Z 2
⋅¿ ¿
] (i)
Now
2
X =X⋅X=
[−11 −11 ]⋅[−11 −11 ]=[−22 −22 ]=2[−11 −11 ]=2 X , (ii)

YX +ZY =[ 1 −1 ] [ 1 −1 ]+ [−2 ] [ 1 −1 ]=[ 2 −2 ] +[−2 2 ] =[ 0 0 ] =O⋅¿ ¿

−1 1 (iii)
Hence (i) gives,

X2 O X2 O X 2 O X4 O
A=
2
[ ]
O Z2
, [ ][ ] [
4
A =( A ) =
22

O Z2 O Z 2
=
O Z4
,
] and
4 4 8
X O X O X O
8 4 2
A =( A ) =
[ ][ ] [ ] ¿ ¿
O Z4 O Z 4 =
O Z8
⋅

Because X 2=2 X , we get

2 8 42 2 2
X 4 =( 2 X ) =4 X 2=4 ( 2 X )=8 X and X =( X ) =( 8 X ) =64 X =64 ( 2 X ) =128 X⋅¿ ¿
8
Clearly, Z 8=(−2) =256⋅¿ ¿ Hence (iv) gives,
1 −1 0
8
A=
128 X O
[
O 256
=128
X O
O 2 ] [ ]
=128 −1 1 0
0 0 2 [ ]
NOTE: Partitioning of a matrix needs lot of experience. One can learn easily with some practice and solving the
problems of exercises.