the Operation

Matrix
Matrix algebra is a means of making
calculations upon arrays of numbers (or data).

a11 ,, a1n 

a 21 ,, a 2n 
A    Aij 
   
 
am1 ,, amn 
Why use it?
 Matrix algebra makes mathematical
expression and computation easier.
 It allows you to get rid of cumbersome
notation, concentrate on the concepts
involved and understand where your results
come from.
Scalar
a scalar is a number

 (denoted with
regular type: 1 or 22)
Row Vector

[1 x n] matrix

A a1 a 2 ,, an  aj

Column Vector
[m x 1] matrix

a1 
a 2 
A     ai
 
 
 am 
Square Matrix
Same number of rows and columns

5 4 7
B  3 6 1 
 
2 1 3 
 The
Identity
Identity Matrix
Square matrix with ones on the
diagonal and zeros elsewhere.

1 0 0 0 
0 1 0 0 
I  
0 0 1 0 
 
0 0 0 1
Transpose Matrix
Rows become columns and
columns become rows

a11 a 21 ,, am1 

a12 a 22 ,, am 2 
A'  
     
 
a1n a 2n ,, amn 
The Transpose of a Matrix

 A is the m×n matrix given by

 a11 a12  a1n 
a a22  a2 n 
A 21 .
    
 
 m1 m 2
a a  a mn 

 The transpose matrix of A, denoted by AT, is a n×m

matrix given by
 a11 a21  am1 
a a  a 
A   12 22
T m2 
,
   
 
 1n 2 n
a a  a mn 
Matrix Addition and
Subtraction

A new matrix C may be defined as the

additive combination of matrices A and B
where: C = A + B
is defined by:

Cij  Aij  Bij

Note: all three matrices are of the same dimension
Addition
a11 a12 
A 
If 
a 21 a 22 


b11 b12 
and B 

b 21 b 22


a11  b11 a12  b12 

then C 

a 21  b 21 a 22  b22 

Matrix Addition Example

3 4  1 2  4 6 

A  B     C
  
5 6  3 4 
 
8 10 

Matrix Subtraction

C = A - B
Is defined by

Cij  Aij  Bij

Matrix Multiplication
Matrices A and B have these dimensions:

[r x c] and [s x d]
Matrix Multiplication
Matrices A and B can be multiplied if:

[r x c] and [s x d]

c=s
Matrix Multiplication

The resulting matrix will have the dimensions:

[r x c] and [s x d]

rxd
Matrix Multiplication
 Thus a11 a12 a13   b11 b12  c11 c12 
a a a  x b b   c c 
 21 22 23   21 22   21 22 
a31 a32 a33  b31 b32  c31 c32 

 where
c11  a11b11  a12b21  a13b31
c12  a11b12  a12b22  a13b32
c 21  a21b11  a22b21  a23b31
c 22  a21b12  a22b22  a23b32
c31  a31b11  a32b21  a33b31
c32  a31b12  a32b22  a33b32
Computation: A x B = C

2 3
1 1 1 
A  1 1  and B 


1 0 2 

 
1 0 
[3 x 2] [2 x 3]
A and B can be multiplied

2 *1  3 *1  5 2 *1  3 * 0  2 2 *1  3 * 2  8 5 2 8
C  1*1  1*1  2 1*1  1* 0  1 1*1  1* 2  3   2 1 3 
1*1  0 *1  1 1*1  0 * 0  1 1*1  0 * 2  1  111 

[3 x 3]
Computation: A x B = C

2 3
1 1 1 
A  1 1  and B 


1 0 2 

 
1 0 
[3 x 2] [2 x 3]
Result is 3 x 3

2 *1  3 *1  5 2 *1  3 * 0  2 2 *1  3 * 2  8 5 2 8
C  1*1  1*1  2 1*1  1* 0  1 1*1  1* 2  3   2 1 3 
1*1  0 *1  1 1*1  0 * 0  1 1*1  0 * 2  1  111 

[3 x 3]
Inversion
Matrix Inversion

1 1
B B  BB  I

Like a reciprocal Like the number one

in scalar math in scalar math
Definition
 The inverse of an n by n matrix A is an n by n
matrix B where,
AB = BA = In.
 Please note: Not all matrices have inverses!
Singular matrices don’t have inverse.
 If a matrix has an inverse, then it is called
invertible.
Definition
If A is a square matrix, and if a matrix B of
the same size can be found such that
AB=BA=I , then A is said to be invertible
and B is called an inverse of A . If no such
matrix B can be found, then A is said to be
singular .

Notation:

B  A1
 Properties of Matrix Inverse

 If A is an invertible matrix then its inverse is unique.

 (A-1)-1 = A.
 (Ak)-1= (A-1)k (we will denote this as A-k )
 (cA)-1 = (1/c)A-1, c ≠ 0.
 ( AT)-1 = (A-1)T.
 If A is an invertible matrix, then the system of equations
Ax = b has a unique solution given by x = A-1b.
The inverse of a 2-by-2 matrix

 For a 2-by-2 matrix,

a b 
A 
c d 

1 1  d  b
A   
ad  bc  c a 
The inverse of a 2 x 2 matrix
 From this we deduced that a 2x2 matrix A is
singular if and only if ad-bc = 0.
 This quantity (ad-bc) has some other useful
properties as well and so is defined to be the
determinant of the matrix A.
Determinants of larger
matrices
 There is no “nice” formula for the inverse of larger than
2x2 matrices.
 We still can define the determinant of a larger square
matrix and it will still have the property that the
determinant of A= 0 if and only if A is singular.
 First we need some terminology.
Minors and cofactors
 If A is a square matrix, then the minor Mij of the element
aij of A is the determinant of the matrix obtained by
deleting the i-th row and the j-th column from A.
 The cofactor Cij = (-1)i+jMij.
Definition of a Determinant
If A is a square matrix of order 2 or greater, then the
determinant of A is the sum of the entries in the first
row of A multiplied by their cofactors. That is,

n
det( A)  A   a1 j C1 j
j 1
Determinant of 2-by-2 Matrix

a b
A 
c d

det( A)  A  ad  bc
Determinant of 3-by-3 Matrix

a b c
A   d e f 
 g h i 

e f d f d e
det( A)  A  a b c
h i g i g h
 aei  afh  bdi  cdh  bfg  ceg
 (aei  bfg  cdh)  ( gec  hfa  idb)
Matrix Inversion
How to calculate the matrix inverse?

1 1
A  adj ( A)
A

What is an adjoint matrix?

Adjoint Matrix – Minors and
Cofactors

 The adjoint matrix of [A], Adj[A] is obtained by taking

the transpose of the cofactor matrix of [A].

 The minor for element aij of matrix [A] is found by

removing the ith row and jth column from [A] and then
calculating the determinant of the remaining matrix.
Matrix Inversion
Consider the following set of linear equations.

 2 x1  4 x2  5 x3  36

-3 x1  5 x2  7 x3  7
5 x  3 x  8 x  -31
 1 2 3

The coefficients can be arranged in a matrix form as shown.

 2 -4 5 
-3 5 7 
  
A  
 5 3 -8
Matrix Inversion
The set of equations in matrix form is:

 2 -4 5   x1   36 
 -3 5 7   x    7 
  2  
 5 3 -8  x3   -31

𝐴 𝑥=𝐵
−1
𝑥= 𝐴 𝐵
Minors
 2 4 5  5 7
 3 5 7 
 M 11 = = -61
 5 3  8 3 -8

 2 4 5 
 3 -3 7
 5 7 
M 12 = = -11
 5 3  8 5 -8

 2 4 5  -3 5
 3 5 7  M 13 = = -34
 5 3
 5 3  8
Minors
 2 4 5 
 3 5  2 -4
 7  M 23   26
 5 3  8 5 3

The resulting matrix of minors is:

 -61 -11 -34 

 17 -41 26 
  
M  
-53 29 -2 
 Cofactors
Cofactors are the signed minors.
i+ j
The cofactor of element aij of matrix [A] is: C ij = (-1) M ij
1+ 1
Therefore C 11 = (-1) M 11
1+ 2
C 12 = (-1) M 12
1+ 3
C 13 = (-1) M 13
 -61 11 -34 
 
The resulting matrix of cofactors is: C   -17 -41 -26 
 -53 -29 -2 
 Adjoint matrix
The adjoint matrix of [A], Adj[A] is obtained by taking the
T
transpose of the cofactor matrix of [A]. adj [A ]= [C ]

 -61 11 -34   -61 -17 -53

 11 -41 -29 

C   -17 -41 -26  adj   
A  
 -53 -29 -2  -34 -26 -2 

2 -4 5
Evaluate the determinant A = -3 5 7 = -336
5 3 -8
 - 1 1
[A ] = adj [A ]
Matrix Inversion A

 -61 -17 -53

1  
 A
1
 11 -41 -29
-336  
-34 -26 -2 

 -61 -17 -53   61 17 53 

 -336 -336   336

-336
 336 336 
 
 A  11 -41 -29 

1 -11 41 29 
 A
1

 -336 -336 -336   336 336 336 
   
 -34 -26 -2   17 13 1 
 -336 -336 -336   168 168 168 
If the original square matrix, A, is given by the following expression:

2 1 0
A   1 2  1
 
 0 1 2 

Then, after augmenting by the

identity, the following is obtained:

2 1 0 1 0 0
A I    1 2 1 0 1 0
 
 0 1 2 0 0 1
By performing elementary row operations on the [AI] matrix until
A reaches reduced row echelon form, the following is the final result:

 3 1 1
1 0 0 4 2 4
 1
I A1   0 1 0
1
1 
0 0 1
2 2
 1 1 3
 4 2 4 

The matrix augmentation can now be undone, which gives the following:

3 1 1
1 0 0 4 2 4
I  0 0
1 1
1 A1  1 
  2 2
1 1 3
0 0 1  4 2 4 