Chapter 2

Matrices
2.1 Operations with Matrices
2.2 Properties of Matrix Operations
2.3 The Inverse of a Matrix

MAT1041 - Chapter 2 1
 2.1 Operations with Matrices
Matrix representations:
 An uppercase case: A, B, C, …

 A representative element enclosed in brackets: [aij], [bij]

 a11 a12  a1n 

a  a2 n 
 A rectangular array of numbers:  21 a22
    
 
am1 am 2  amn 

Vector (column/row matrix): boldface lowercase a1, a2, …, an

MAT1041 - Chapter 2 2-2
Section 2-1

Definitions
 Equality of Matrices
Two matrices A = [aij] and B = [bij] are equal if they have the same
size (mn) and aij = bij for 1  i  m and 1  j  n.
 Matrix Addition
If A = [aij] and B = [bij] are matrices of size mn, then their sum is
the mn matrix given by A+B = [aij + bij].
The sum of two matrices of different sizes is undefined.
 Scalar Multiplication
If A = [aij] is an mn matrix and c is a scalar, then the scalar
multiplication of A by c is the mn matrix given by cA = [caij]

MAT1041 - Chapter 2 2-3

Section 2-1

Example 1
Consider the four matrices
1 2  1 1 2
A  , B    , C  1 3, D   
 3 4   
3  x 4 
 Matrices A and B are not equal because they are of different sizes.
 Similarly, B and C are not equal.
 Matrices A and D are equal if and only if (iff) x = 3.

Remark: “p if and only if q” means that p implies q and q implies p.

MAT1041 - Chapter 2 2-4

Section 2-1

Subtraction of Matrices
 If A and B are of the same size, AB represents the sum of A and (B).
That is, AB = A+(1)B = [aij  bij].
 cA  dB = [caij  dbij].
1 2 4 2 0 0
  B   1  4 3
 Example 3: A   3 0  1 and
 2 1 2   1 3 2
1 2 4 2 0 0   3 1  2 3 2  0 3 4  0 
3 A  B  3 3 0  1   1  4 3   3  (3)  1 3  0  ( 4) 3  (1)  3
 2 1 2   1 3 2 3  2  (1) 3 1  3 3  2  2 
 1 6 12 
  10 4  6
 7 0 4  MAT1041 - Chapter 2 2-5
Section 2-1

Matrix Multiplication
 If A = [aij] is an mn matrix and B = [bij] is an np matrix, then the
product AB is an mp matrix AB = [cij], where
n
cij   aik bkj  ai1b1 j  ai 2 b2 j    ainbnj
k 1
 a11 a12  a1n   c11 c12  c1 j  c1 p 
a a22  a2 n  b11 b12  b1 j  b1 p   c21 c22  c2 j  c2 p 
 21 
    b21 b22  b2 j  b2 p      
   
 ai1 ai 2  ain        ci1 ci 2  cij  cip 
 
    bn1 bn 2  bnj  bnp      
   
am1 am 2  amn  cm1 cm 2  cmj  cmp 
A B  AB
mn n p m p
MAT1041 - Chapter 2 2-6
Section 2-1

Example 4
 1 3 
    3 2
Find the product AB, where A  4  2 and B   
    4 1 
 5 0 
 1 3   c11 c12    9 1 
 4  2    3 2   c   4 6
   4 1   21 22   c 
 5 0    c31 c32   15 10
3 2 2 2 3 2
c11  (  1)(  3)  (3)(  4 )   9 c12  (  1)( 2)  (3)(1)  1
c21  (4)(3)  (2)(4)  4 c22  (4)(2)  (2)(1)  6
c31  (5)(3)  (0)(4)  15 c32  (5)(2)  (0)(1)  10

MAT1041 - Chapter 2 2-7

Section 2-1

Example 5
 2 4 2
1 0 3    5 7  1
(a )    1 0 0 
2  1  2   3 6 6  23
  1 
1  1
1 2  1 2  1 0
(c )      0 1 
1 1   1  1   22
2
(d ) 1  2  3 1  1 11 AB  BA
 1 
Matrix multiplication is not,
2  2  4  6 in general, commutative.
   1 2 
(e)  1 1  2  3   3 
 1   1  2  3 33
MAT1041 - Chapter 2 2-8
Section 2-1

Systems of Linear Equations

 Matrix Equation: Ax = b
 a11 a12 a13   x1   b1  a11 x1  a12 x2  a13 x3  b1
a      
 21 a22 a23   x2   b2   a21 x1  a22 x2  a23 x3  b2
 a31 a32 a33   x3  b3  a x  a x  a x  b
 31 1 32 2 33 3 3
 A: coefficient matrix; x and b: column matrix (vector)

 Example 6: Solve the matrix equation Ax = 0, where

 x1   x1  1
1  2 1  0  1 0  17 0
A  
, x   x2  , 0      x   x2   t 4, t  R.
 2 3  2 
  
0  0 1  4
0
 x3  7
 x3  7

MAT1041 - Chapter 2 2-9

Section 2-1

Diagonal Matrix & Trace (p. 58)

 A square matrix A is called a diagonal matrix if all entries that not
on the main diagonal are zero.

 The trace of an nn matrix A is the sum of the main diagonal

entries. That is,
Tr ( A)  a11  a 22    a nn

MAT1041 - Chapter 2 2-10

 2.2 Properties of Matrix Operations
Theorem 2.1 Properties of Matrix Addition and Scalar Multiplication
 If A, B, and C are mn matrices and c and d are scalars, then the
following properties are true.
1. A+B = B+A Commutative property of addition
2. A+(B+C) = (A+B)+C Associative property of addition
3. (cd)A = c(dA) Associative property of multiplication
4. 1A = A Multiplication identity
5. c(A+B) = cA + cB Distributive property
6. (c+d)A = cA + dA Distributive property

MAT1041 - Chapter 2 2-11

Section 2-2

Proof of Theorem 2.1

 The proofs follow directly from the definitions of matrix addition
and scalar multiplication, and the corresponding properties of real
numbers.
 Let A = [aij] and B = [bij]

1. Use the commutative properties of addition of real numbers to write

A+B = [aij+bij] = [bij+aij] = B+A
5. Use the distributive properties (for real number) of multiplication
over addition to write
c(A+B) = [c(aij+bij)] = [caij+cbij] = cA+cB

MAT1041 - Chapter 2 2-12

Section 2-2

Zero Matrix & Additive Identity

 If A is an mn matrix and Omn is the mn matrix consisting entirely
of zeros, then A + Omn = A.
 The matrix Omn is called a zero matrix, and it serves as the additive
identity for the set of all mn matrices.

 Theorem 2.2: Properties of Zero Matrix

If A is an mn matrix and c is a scalar, then the following properties
are true.
1. A + Omn = A.
2. A + (A) = Omn.  A is the additive inverse of A.
3. If cA = Omn, then c = 0 or A = Omn.
MAT1041 - Chapter 2 2-13
Section 2-2

Matrix Equation
Real Numbers m  n Matrices
xa b X  A B
x  a  (a)  b  (a) X  A  ( A)  B  ( A)
x0ba X O  B A
x ba X  B A
Exercise 2: Solve for X in the equation 3X+A = B, where
1  2   3 4
A  and B   
 0 3   2 1 
1 1   3 4 1  2  1  4 6   43 2
X  ( B  A)        2
3    
3   2 1 0 3   3  2  2  3  23 
MAT1041 - Chapter 2 2-14
Section 2-2

Theorem 2.3
 Properties of Matrix Multiplication
If A, B, and C are matrices (with sizes such that the given matrix products
are defined) and c is a scalar, then the following properties are true.
1. A(BC) = (AB)C Associative property
2. A(B+C) = AB + AC Distribution property
3. (A+B)C = AC + BC Distribution property
4. c(AB) = (cA)B = A(cB)
Proof of Property 2: A: mn matrix, B: np matrix, C: np matrix.
The entry in the ith row and jth column of A(B+C) is
a i1 (b1 j  c1 j )  a i 2 (b2 j  c 2 j )    a in (bnj  c nj )
The entry in the ith row and jth column of AB +AC is equal
( a i1b1 j  a i 2 b2 j    a in bnj )  ( a i1c1 j  a i 2 c 2 j    a in c nj )

MAT1041 - Chapter 2 2-15

Section 2-2

Noncommutativity
 A commutative property for matrix multiplication was NOT listed
in Theorem 2.3.
 If A is of size 23 and B is of size 33,
then the product AB is defined, but the product BA is not.
Example 4: Show that AB and BA are not equal for the matrices
1 3  2  1
A  B 
 2  1  0 2 
1 3  2  1 2 5 
AB       
2  1 0 2  4  4 AB  BA
2  1 1 3  0 7 
BA       
 0 2  2  1  4  2 
MAT1041 - Chapter 2 2-16
Section 2-2

Cancellation Property
 It does NOT have a general cancellation property for matrix
multiplication.
 If AC = BC, it is NOT necessary true that A = B.

Example 5: Show that AC = BC.

 1 3 2 4  1  2
A  , B  , C 
 0 1   2 3    1 2 
1 3 2 4  2 4
AB   
0 1 2 3   1 2
  
AC  BC , but A  B
1 3  1  2  2 4
AC      
0 1   1 2    1 2
MAT1041 - Chapter 2 2-17
Section 2-2

Identity Matrix & Theorem 4

 A square matrix that has 1’s on the main diagonal and 0’s elsewhere.
1 0 0  0
0 1 0  0

 The identity matrix of order n: I n  0 0 1  0
 
   
0 0 0  1

Theorem 2.4: Properties of the Identity Matrix

If A is a matrix of size mn, then the following properties are true.
1. AIn = A. 2. ImA = A.

 If A is a square matrix of order n, then AIn = InA = A.

MAT1041 - Chapter 2 2-18
Section 2-2

Repeated Multiplication
 Repeated multiplication of a square matrix:
For a positive integer k, Ak is A k  AA  A
k factors
 A0 = In, where A is a square matrix of order n.
 A j A k  A j  k j and k are nonnegative integer.
 ( A j ) k  A jk
2  1
Example 3: Find A for the matrix A  
3
  .
3 0 
  2  1  2  1   2  1 1  2  2  1   4  1 
A   
3
   
       
 3 0  3 0   3 0   6  3  3 0   3  6 
MAT1041 - Chapter 2 2-19
Section 2-2

Theorem2.5 Number of Solutions of a System of Linear Equations

 For a system of linear equations in n variables, precisely one of the
following is true.
1. The system has exactly one solution.
2. The system has an infinite number of solutions.
3. The system has no solution.

MAT1041 - Chapter 2 2-20

Section 2-2

The Transpose of a Matrix

 The transpose of a matrix is formed by writing its columns as rows.
 a11 a12 a13  a1n   a11 a21 a31  am1 
a a am 2 
 21 a22 a23  a2 n   12 a22 a32 
A   a31 a33  a3n   A   a13 a23 a33  am 3 
T
a32
   
           
am1 am 2 am 3  amn  mn a1n a2 n a3n  amn  nm

 A matrix A is symmetric if A = AT.  aij = aji,  i  j.

 a symmetric matrix must be square.

MAT1041 - Chapter 2 2-21

Section 2-2

Theorem 2.6 Properties of Transpose

If A and B are matrices (with sizes such that the given matrix products
are defined) and c is a scalar, then the following properties are true.
1. ( AT )T  A Transpose of a transpose
2. ( A  B)T  AT  BT Transpose of a sum
3. (cA)T  c( AT ) Transpose of a scalar multiplication
4. ( AB ) T  B T AT Transpose of a product

 ( A  B  C )T  AT  BT  C T ( ABC ) T  C T B T A T
 For any matrix A, the matrix AAT is symmetric.
pf : ( AAT )T  ( AT )T AT  AAT
MAT1041 - Chapter 2 2-22
Section 2-2

Example 9
 Show that ( AB ) T  B T AT are equal.
2 1  2 3 1 
A   1 0 3  B  2  1
 0  2 1  3 0 
Sol: 2 1  2 3 1   2 1 
       2 6  1
AB   1 0 3  2  1   6  1  ( AB)  
T

 1  1 2 
 0  2 1  3 0   1 2 
 2 1 0  ( AB ) T  B T AT
3 2 3   2 6  1
B A 
T T
 0  2  
1  1 2 
 1
1  1 0  2 3 1  

MAT1041 - Chapter 2 2-23
Section 2-2

Example 10
T
 Find the product AA and show that it is symmetric.
1 3
A   0  2
Sol:  2  1 
1 3  10  6  5
1 0  2 
AAT   0  2  
   6 4 2  a a
 ij ji
 
 2  1     5 2
3 2 1
 3 

Since AA  ( AA ) , AAT is symmetric.

T T T

MAT1041 - Chapter 2 2-24

 2.3 The Inverse of a Matrix
 Definition of an Inverse of a Matrix
An nn matrix A is invertible (or nonsingular) if there exists an nn
matrix B such that AB = BA = In
 In is the identity matrix of order n.
 The matrix B is called the (multiplicative) inverse of A.
 A matrix that does NOT have an inverse is called noninvertible (or singular).
 Nonsquare matrices do NOT have inverse.

MAT1041 - Chapter 2 2-25

Section 2-3

Theorem 2.7 Uniqueness of an Inverse Matrix

If A is an invertible matrix, then its inverse is unique.
The inverse of A is denoted by A 1 .
AA  A  1 A  I
1
Proof:
 Because A is invertible, it has at least one inverse B such that AB = BA = I.

 Suppose that A has another inverse C such that AC = CA = I.

 Then you can show that B and C are equal as follows.

AB = I
 C(AB) = C(I)
 (CA)B = C
 (I)B = C  B = C
 Consequently B = C, and it follows that the inverse of a matrix is unique.

MAT1041 - Chapter 2 2-26

Section 2-3

Example 2
1 4
Find the inverse of the matrix A   
  1  3
Sol: To find the inverse of A, try to solve the matrix equation AX = I for X.
1 4   x11 x12  1 0
 1  3  x    1  3  4
   21 22  
x 0 1  A X  
 1 1 
 x11  4 x21 x12  4 x22  1 0
    Using matrix multiplication
  x11  3 x 21  x 12  3 x 22   0 1  to check the result.
 x11  4 x21  1
  x11  3, x21  1 1 4    3  4  1 0 
 x11  3 x21  0   1  3  1 
  1  0 1 

 x12  4 x22  0
  x12  4, x22  1
 x12  3 x22  1
MAT1041 - Chapter 2 2-27
Section 2-3

Gauss-Jordan Elimination
 x11  4 x21  1  x12  4 x22  0
 
 x11  3 x21  0  x12  3 x22  1
1 4 1 1 4 0
 1  3 0   1  3 1   The same coefficient matrix
   
1 4 1 0 1 4 1 0 1 0  3  4
  0 1 1 
 1  3 0 1 
  0 1 1 1 (4)  1 

[ A ┇ I ]  …  [ I ┇ A1 ]
Double augment matrix

MAT1041 - Chapter 2 2-28

Section 2-3

Procedure to find the inverse

Let A be a square matrix of order n.
 Write the n2n matrix [ A┇I ] (adjoining the matrices A & I)

 If possible, row reduce A to I using elementary row operations on

the entire matrix [ A┇I ].

The result will be the matrix [I ┇A1 ].
If this in not possible, then A is not invertible.
1
 Check your work by multiplying to see that AA  I  A 1 A

MAT1041 - Chapter 2 2-29

Section 2-3

Example 4
1 2 0
 Show that the matrix A has no inverse. A 
 3 1 2 
 
Sol:  2 3  2
 1 2 0 1 0 0 (3)
A I    3  1 2 0 1 0 2
 2 3  2 0 0 1
1 2 0 1 0 0
 0  7 2  3 1 0
0 7 2 2 0 1
It is not possible to rewrite
1 2 0 1 0 0 [A┇I ] in the form [I ┇A1 ].
 0  7 2  3 1 0
Hence A has no inverse.
0 0 0  1 1 1
MAT1041 - Chapter 2 2-30
Section 2-3

The Inverse of a 22 Matrix


a b 
 The matrix A is a 22 matrix given by A   
 c d 
+
 The matrix A is invertible if and only if  = ad  bc  0.

1 1  d  b Interchanging the entires on the

 If   0, then 
   c a 
A main diagonal and changing the
signs of the other two entires.
1 1 a b   d  b 
Proof: AA 
ad  bc  c d   c a 
1 ad  bc 0  1 0
  
ad  bc  0 ad  bc  0 1


MAT1041 - Chapter 2 2-31

Section 2-3

Example 5
 If possible, find the inverse of each matrix.
 3  1
(a) A   
  2 2 
1 1 2 1  12 1

  ad  bc  4  0  A    1 4
 
4 2 3  2 3
4 
 3  1
(b ) B   
  6 2 
  ad  bc  (3)(2)  (1)(6)  0
The matrix B is not invertible.

MAT1041 - Chapter 2 2-32

Section 2-3

Theorem 2.8 Properties of Inverse Matrix

If A is an invertible matrix, k is a positive integer, and c is a scalar,
then A1, Ak, cA, and AT are invertible and the flowing are true.
1. ( A 1 ) 1  A 2. ( A k ) 1  ( A 1 ) k
3. ( cA ) 1  1c A 1 , c  0 4. ( AT ) 1  ( A 1 ) T

Hint: if BC = CB = I, then C is the inverse of B.

Proof: 1. Observe that AA 1  A 1 A  I , which means that A is the inverse of A1.
Thus, ( A 1 ) 1  A .

 1
 1
)  (1) I  I Hence C1 A 1 is the inverse of (cA),
3. 
1 1
( cA)( c A ) ( c c )( AA
 1 1 1
1 1
which implies that ( cA )  1c A , c  0

 c
( A )(cA )  ( 1
c c )( A A)  (1) I  I

MAT1041 - Chapter 2 2-33

Section 2-3

Example 6
Compute A2 in two different ways and show that the results are equal.
1 1 
A 
2 1  2 4  1 2
1. (A ) : 2. (A ) :
1 1  1 1   3 5  1 1 
A 
2
     A     (1)( 4)  ( 2)(1)  2
 2 4  2 4   10 18  2 4
   (3)(18)  (5)(10)  4 1 1  4  1  2  12 
A     1
   54  2  2 1  1 2
1  18 5  9
 (A2 ) 1    
2
3
4  10 3  52 4
 2  12   2  12   92
1 2
(A )    5
 54 
1  1 3
 1 2   1 2  2 4
the same result

MAT1041 - Chapter 2 2-34

Section 2-3

Theorem 2.9 The Inverse of a Product

If A and B are invertible matrices of size n, then AB is invertible and
(AB)1 = B1A1.

Proof:
1. (AB)(B1A1) = A(BB1)A1 = A(I)A1 = AA1 = I.
2. (B1A1)(AB) = B1(A1A)B = B1(I)B = B1B = I.
Hence AB is invertible.

 ( A1 A2  An ) 1  An1  A21 A11 : in reverse order

Recall: (AB)T = BTAT
MAT1041 - Chapter 2 2-35
Section 2-3

Example 7
1 3 3 1 2 3

Find (AB)1 for the matrices A  1 4 3 and B  1 3 3 using the fact that
  
A1 and B1 are given by 1 3 4 2 4 3
 7  3  3  1 2 1
 0
A1   1 0 B   1
1
1 1
Sol: 
  1 0 1 
 

2
3 0  13 

 1 2 1  7  3  3  8  5  2
( AB) 1  B 1 A 1    1 1 0  1 1 0   8 4 3
 23 0  13   1 0 1  5  2  73 
10 23 21 10 23 21 1 0 0  1 0 0 8  5  2
AB  11 26 24  11 26 24 0 1 0  0 1 0  8 4 3 
   
12 27 24 12 27 24 0 0 1 0 0 1 5  2  3 
7

MAT1041 - Chapter 2 2-36

Section 2-3

Theorem 2.10 Cancellation Property

If C is an invertible matrix, then the following properties hold.
1. If AC = BC, then A = B. Right cancellation property
2. If CA = CB, then A = B. Left cancellation property
Proof: Use that fact that C is invertible and write
AC = BC
 (AC)C1 = (BC)C1
 A(CC1) = B(CC1)
 AI = BI
A=B

MAT1041 - Chapter 2 2-37

Section 2-3

Theorem2.11 Systems of Equations with Unique Solutions

If A is an invertible matrix, then the system of linear equations Ax = b
has a unique solution given by x = A1b.
Proof: Ax = b  A1(Ax) = A1(b)  A1Ax = A1b  x = A1b

Example 8: Use an inverse matrix to solve each system

( a ) 2 x  3 y  z  1 2 3 1  1 1 0
3x  3 y  z  1 A  3 3 1  A1   1 0 1
   
2 x  4 y  z  1 2 4 1  6  2  3

(c ) 2 x  3 y  z  0  1 1 0   1  2 0 
3x  3 y  z  0 ( a ) x  A b   1
1
0 1  1    1 (c) x  A1b  0
      
2x  4 y  z  0  6  2  3  2  2 0
MAT1041 - Chapter 2 2-38