Advanced Mathematics 2 - Linear Algebra

Chapter 2: Matrix algebra

Department of Mathematics
The FPT university

TrungDT (FUHN) MAA101 Chapter 2 1 / 29

Chapter 2 Introduction

Topics:

2.1 Matrix operations: Addition, Scalar multiplication, Transposition

2.2 Matrix multiplication
2.3 Matrix inverses
2.5 Matrix transformations
2.7 Applications

TrungDT (FUHN) MAA101 Chapter 2 2 / 29

2.1 Matrix addition, Scalar multiplication, and
Transposition

A matrix is an array consisting of m rows and n columns:

 
a11 a12 · · · a1n
 a21 a22 · · · a2n 
A= · · ·

· · · aij · · · 
am1 am2 · · · amn

We say A is an m × n matrix, or A is a matrix of size m × n. We

usually write A = [aij ], i = 1, . . . , m, j = 1, . . . , n.
The number aij is called the (i, j)-entry of the matrix.
If m = n then we say A is a square matrix.

TrungDT (FUHN) MAA101 Chapter 2 3 / 29

Matrix Addition and Scalar multiplication

Addition and Subtraction

     
a b c A B C a±A b±B c ±C
± =
x y z X Y Z x ±X y ±Y z ±Z
   
a b c A B
+ = Undefined
x y z X Y
Scalar multiplication
   
a b c ka kb kc
k = , where k is a real number
x y z kx ky kz
   
1 −2 −1 0
Example. Find X : 2X + =
3 4 3 1

TrungDT (FUHN) MAA101 Chapter 2 4 / 29

Let A, B, C be m × n matrices. Let 0 denote the m × n matrix whose
entries are 0. Let k, p be real numbers.
Theorem
A+B =B +A
A+0=0+A=A
A + (B + C ) = (A + B) + C
k(A + B) = kA + kB
kA + pA = (k + p)A
k(pA) = (kp)A
1.A = A

TrungDT (FUHN) MAA101 Chapter 2 5 / 29

Transpose
The transpose of the matrix A, denoted by AT , is the matrix whose
ith-row is the ith-column of A.
 
 T 1 x
1 2 3
Example. = 2 y 
x y z
3 z
Properties
(AT )T = A
(A + B)T = AT + B T
(kA)T = kAT

Example. Find A such that:

   

T 1 2 T 1 −1
2A + =
3 4 2 3

TrungDT (FUHN) MAA101 Chapter 2 6 / 29

Symmetric matrices
A square matrix A is symmetric if A = AT .
Equivalently, a square matrix A is symmetric if and only if the entries that
are symmetric across the main diagonal are equal.
  
1 3 1 3
Example. The matrix is symmetric, but is not.
3 5 2 5

Question. Is the sum of two symmetric matrices symmetric?

TrungDT (FUHN) MAA101 Chapter 2 7 / 29

2.2 Matrix multiplication

Let A be a m × n matrix. Let B be an n × p matrix. Then the product of

A and B, denoted by AB, is a matrix of size m × p.

Examples.
 
  M N  
a b c  aM + bP + cU aN + bQ + cV
P Q =
x y z xM + yP + zU xN + yQ + zV
U V
 
  a  
2 3 4 b  = 2a + 3b + 4c
c
   
2   2a 2b 2c
3 a b c = 3a 3b 3c 
4 4a 4b 4c

TrungDT (FUHN) MAA101 Chapter 2 8 / 29

The identity matrix In is the n × n matrix with 1 on the diagonal and 0
everywhere else.
Theorem
Let A, B, C be matrices of sizes such that the indicated operations can be
performed. I denotes an identity matrix, and k a real number.
AI = IA = A
AB 6= BA, in general
A(BC ) = (AB)C
A(B + C ) = AB + AC
(B + C )A = BA + CA
k(AB) = (kA)B = A(kB)
(AB)T = B T AT

TrungDT (FUHN) MAA101 Chapter 2 9 / 29

Block multiplication

To multiply two matrices of large sizes, we can divide each matrix into
blocks and treat these as entries.

Example 1.
     
1 2 0 0 9 10 1 2 0 0 9 10
3 4 0 0 11 12  3 4 0 0
  11 12 
 
1 0 5 6  0 0  =  1 0
   
6 5 0 0 
0 1 7 8 0 0 0 1 8 7 0 0
 
     31 34
A 0 C AC 71 78
= = = 
I B 0 C  9 10
11 12

TrungDT (FUHN) MAA101 Chapter 2 10 / 29

Example 2.
 10  
1 0 1 2  10   1 0 10 20
 0 1 3 4  I A I 10A 0 1 30 40
  = = = 
 0 0 1 0  0 I 0 I 0 0 1 0
0 0 0 1 0 0 0 1

TrungDT (FUHN) MAA101 Chapter 2 11 / 29

Matrices and Linear systems

A system of linear equations



 a11 x1 + a12 x2 + · · · + a1n xn = b1
a21 x1 + a22 x2 + · · · + a2n xn = b2


 ···
am1 x1 + am2 x2 + · · · + amn xn = bm


can be written as a matrix equation AX = B, where

   

a11 a12 ··· a1n
 x1 b1
x2   b2 
 a21 a22 ··· a2n 
A= , X =  . , B =  . 
   
 ···   ..   .. 
am1 am2 ··· amn xn bm

TrungDT (FUHN) MAA101 Chapter 2 12 / 29

2.3 Matrix inverses

Let A be an n × n matrix. An n × n matrix B is called an inverse of A if

AB = BA = In

Notes.
 
1 2
A matrix may not have an inverse. For example, the matrix
3 6
does not have an inverse.
The inverse matrix of a matrix A, if exists, is unique and is denoted
by A−1 . In this case we say A is invertible.

TrungDT (FUHN) MAA101 Chapter 2 13 / 29

Inverses of 2 × 2 matrices
 
a b
Let A = . The determinant of A = det(A) := ad − bc.
c d

Theorem
 
a b
The matrix A = is invertible if and only if det(A) 6= 0. In that case,
c d
 
1 d −b
A−1 = .
det(A) −c a

Example. Find inverses of the matrices

   
1 2 2 4
A= , B=
3 4 3 6

TrungDT (FUHN) MAA101 Chapter 2 14 / 29

Application in solving linear systems

Consider a system of linear equations AX = B. If the matrix A is

invertible, then X = A−1 B is the solution.

Example. Solve the system


2x1 − x2 = 3
3x1 + 4x2 = −1

TrungDT (FUHN) MAA101 Chapter 2 15 / 29

Matrix Inversion Algorithm
reduced
Let A be invertible. Then A =⇒ I , and with the same operations
reduced
I =⇒ A−1 .
reduced-row echelon
To find A−1 , perform [A|I ] =⇒ [I |B]. Then B is the
inverse of A.
 
2 7 1
Example. Find the inverse of the matrix A = 1 4 −1.
1 3 0
   
2 7 1 1 0 0 1 0 0 −3/2 −3/2 11/2
 1 4 −1 0 1 0  reduced =⇒  0 1 0 1/2 1/2 −3/2 
1 3 0 0 0 1 0 0 1 1/2 −1/2 −1/2
 
−3/2 −3/2 11/2
Then A−1 =  1/2 1/2 −3/2 
1/2 −1/2 −1/2
TrungDT (FUHN) MAA101 Chapter 2 16 / 29
Theorem
I −1 = I , where I is an identity matrix.
(A−1 )−1 = A
If A and B are invertible then (AB)−1 = B −1 A−1 .
If A is invertible then (An )−1 = (A−1 )n .
If A is invertible and c 6= 0 then (cA)−1 = c1 A−1
If A is invertible then (AT )−1 = (A−1 )T .

Example. Find the matrix A that satisfy:

 
T −1 2 1
(A − 2I ) =
−1 0

TrungDT (FUHN) MAA101 Chapter 2 17 / 29

Theorem
Let A be an n × n matrix. The following are equivalent:
A is invertible.
The homogeneous system AX = 0 has only the trivial solution X = 0.
reduced
A =⇒ I .
The system AX = B has at least one solution for every choice of
column B.
There is an n × n matrix C such that AC = I .

TrungDT (FUHN) MAA101 Chapter 2 18 / 29

2.5 Matrix transformation

T is reflection in the x-axis

y
      
  x x 1 0 x
6 x T = =
*• y
 y −y 0 −1 y

 - x
o H
H  
x 1 0
 
j• −y
H
H T is induced by the matrix
0 −1
Trên là reflect qua Ox,
Reflect qua Oy, T is induced by |-1 0|
| 0 1|

Question. Find the matrix of the following transformation in the plane:

Reflection là đối xứng qua trục, nên ý a) là
(a) Reflection in the y -axis đối xứng qua trục Oy, tức là điểm [-x,y]
(b) Projection on the x-axis Projection là chiếu, nên ý b) là chiếu trên
trục Ox, tức là điểm [x,0]
TrungDT (FUHN) MAA101 Chapter 2 19 / 29
y
T is rotation through an angle θ
6 x0
 
• 0    0   
 y   x x cos θ − sin θ x
x T = 0 =

*• y
 y y sin θ cos θ y
 
θ
o - x  
cos θ − sin θ
T is induced by the matrix
sin θ cos θ

TrungDT (FUHN) MAA101 Chapter 2 20 / 29

 A map T : Rn → Rm is a matrix transformation if there is a matrix A
such that T (~v ) = A~v for all ~v ∈ Rn .
This matrix A is of size m × n. We say the matrix of T is A, or T is
induced by the matrix A.

Example. T : R3 → R2 , T [x, y , z]T = [2x T

 + y , x − y + z] , is a matrix
2 1 0
transformation; it is induced by the matrix
1 −1 1

TrungDT (FUHN) MAA101 Chapter 2 21 / 29

Linear transformations

A map T : Rn → Rm is a linear transformation if for all ~u , ~v in Rn :

T (~u + ~v ) = T (~u ) + T (~v )
T (k~u ) = kT (~u ), where k is a real number

Example. Check if the transformation is linear:

 
a
(a) T : R2→ R, T = ab
b
 
2 a
(b) T : R → R, T = a + 2b
b
 
a
(c) T : R2 → R, T = a + 2b + 1
b

TrungDT (FUHN) MAA101 Chapter 2 22 / 29

Property of Linear Transformation
Let T : Rn → Rm be a linear transformation. Then

x2 + · · · + ak →
x1 + a2 →
T (a1 →
− − −
xk ) = a1 T (→
−
x1 ) + a2 T (→
−
x2 ) + · · · + ak T (→
−
xk )

for all ai ∈ R and →

−
xi ∈ Rn .

Example 1. Let T : R2 → R2 be a linear transformation satisfying

   
−2 1
T (u) = , T (v ) = .
3 −1
Find T (2u − 5v ).

Example 2. Let T : R2 → R2 be a linear transformation satisfying

       
1 2 1 5
T = ,T = .
1 −3 −2 1
 
4
Find T .
3
TrungDT (FUHN) MAA101 Chapter 2 23 / 29
Linear Transformation and Matrix Linear Transformation

Theorem
Let T : Rn → Rm be a linear transformation. Then T is a matrix
transformation induced by the following m × n matrix

A = [T (E1 ) T (E2 ) . . . T (En )]

     
1 0 0
0 1 0
where E1 =  .  , E2 =  .  , . . . , En =  . .
     
 ..   ..   .. 
0 0 1

TrungDT (FUHN) MAA101 Chapter 2 24 / 29

Composite of linear transformations

The composite of S and T , denoted by S ◦ T , is a transformation from Rn

to Rp . It is defined by

S ◦ T (→
−
x ) = S(T (→
−
x ))

for all →
−
x ∈ Rn .
TrungDT (FUHN) MAA101 Chapter 2 25 / 29
Theorem
If the matrix of T is A and the matrix of S is B then the matrix of S ◦ T
is BA.

Example. Let T be reflection in the x-axis followed by rotation through

π/2. Find the matrix of T .
 
0 −1
The matrix of rotation Rπ/2 through π/2 is A = .
1 0
 
1 0
The matrix of reflection S in x-axis is B = .
0 −1

The matrix of the composite T = Rπ/2 ◦ S is

    
0 −1 1 0 0 1
AB = =
1 0 0 −1 1 0

TrungDT (FUHN) MAA101 Chapter 2 26 / 29

2.7 Application: Input-Output economic models

A society has three basic needs: food, shelter, clothing.

There are three industries in the society - farming, housing, garment -
that produce these commodities.
Each of these industries consumes a certain proportion of the total
output of each commodity.

OUTPUT
Farming Housing Garment
Farming 0.4 0.2 0.3
CONSUMPTION Housing 0.2 0.6 0.4
Garment 0.4 0.2 0.3

Problem. Find the annual prices that each industry must charge for its
income to equal its expenditures.
TrungDT (FUHN) MAA101 Chapter 2 27 / 29
 OUTPUT
Farming Housing Garment
Farming 0.4 0.2 0.3
CONSUMPTION Housing 0.2 0.6 0.4
Garment 0.4 0.2 0.3

Let p1 , p2 , p3 be the annual prices charged by the farming, housing and

garment industries.

 0.4p1 + 0.2p2 + 0.3p3 = p1
0.2p1 + 0.6p2 + 0.4p3 = p2
0.4p1 + 0.2p2 + 0.3p3 = p3


Solution: (p1 , p2 , p3 ) = (2t, 3t, 2t)

TrungDT (FUHN) MAA101 Chapter 2 28 / 29

Put    
0.4 0.2 0.3 p1
E = 0.2 0.6 0.4 , P = p2 
  
0.4 0.2 0.3 p3

Then the equations can be rewritten as

(I − E )P = 0

The matrix E is called the input-output matrix for the economy.

Solutions P to the equation (I − E )P = 0 are called equilibrium price

structures.

TrungDT (FUHN) MAA101 Chapter 2 29 / 29