ch27 Matrices and Applications

P1: FXS/ABE P2: FXS
9780521740517c27.xml CUAU031-EVANS September 7, 2008 7:15
Back to Menu >>>
C H A P T E R
27
MODULE 6
Matrices and
E
applications II
PL
What is an identity matrix?
What is the inverse of a matrix and how is it determined?
How do we use inverse matrices to solve sets of simultaneous equations?
How do we define the power of a matrix?
What is a transition matrix and how is it used?
The inverse matrix

M
27.1
The identity matrix
Before we introduce the inverse of a matrix, we need to introduce the identity matrix. In
chapter 26, we met the zero matrix, a matrix of all zeros. In the arithmetic of matrices, this
plays the same role as one plays in normal arithmetic. In matrix arithmetic, the identity matrix
plays the same role as the number 1 in normal arithmetic. It is only defined for square matrices.
SA
The identity matrix

A square matrix with ones on the leading diagonal and zeros elsewhere is known as an
identity matrix. The symbol I is used to represent the identity matrix. The matrices below
are all examples of identity matrices.
 
  1 0 0 0
1 0 0 0 1 0 0
1 0    
I = , I = 0 1 0, I =  
0 1 0 0 1 0
0 0 1
0 0 0 1
An identity matrix has the property that, for a square matrix A, IA = AI = A.
Just as the product of a number and one, is the same number (1 × 3 = 3 × 1 = 3), the product
of a square matrix and the identity matrix is the same matrix, IA = AI = A.
Cambridge University Press • Uncorrected Sample pages • 978-0-521-61328-6 • 2008 © Jones, Evans, Lipson
722 and McMenamin
TI-Nspire & Casio ClassPad material in collaboration with Brown
P1: FXS/ABE P2: FXS
Back to Menu >>>

Chapter 27 — Matrices and applications II 723
Example 1 Verifying that IA = AI = A

1 2
For the matrix A = , show that IA = AI = A.
3 4
Solution
1 For a 2× 2 matrix, the identity matrix
1 2 1 0
1 0 A = I=
is I = . Write down A and I. 3 4 0 1
0 1
E

2 Form the product AI and evaluate. 1 2 1 0
Use a calculator if you wish. AI =
3 4 0 1

1×1+2×0 1×0+2×1 1 2
= =
3×1+4×0 3×0+4×1
PL
3 Form the product IA and evaluate.
Use a calculator if you wish.
∴ AI = A
IA =
=

1 0
0 1

1 2
3 4

1×1+0×3 1×2+0×4
0×1+1×3 0×2+1×4
∴ IA = A

=
1
3
3 4
2
4

M
The inverse matrix A−1
Having defined the properties of the identity matrix, we are now in the position to define a new
matrix concept, the inverse matrix.
The inverse matrix A−1

SA
The inverse of a square matrix A, is called A−1 and has the property that
A A−1 = A−1 A = I .
Before we investigate how to calculate an inverse matrix, we need to show you that they
actually exist.
Example 2 Verifying that AA-1 = A-1 A = I

2 3 −1 5 −3
The matrix A = has an inverse A = . Show that A A−1 = A−1 A = I .
3 5 −3 2
Solution
2 3 −1 5 −3
1 Write down A and A−1 . A= A =
3 5 −3 2
TI-Nspire & Casio ClassPad material in collaboration with Brown and McMenamin
P1: FXS/ABE P2: FXS
Back to Menu >>>

724 Essential Further Mathematics — Module 6 Matrices and applications

2 Form the product A A−1 2 3 5 −3
and evaluate. AA−1 =
3 5 −3 2
Use a calculator if you
2 × 5 + 3 × (−3) 2 × (−3) + 3 × 2 1 0
wish. = =
3 × 5 + (−3) × 5 3 × (−3) + 5 × 2 0 1
AA−1 = I

3 Form the product A−1 A 5 −3 2 3
and evaluate. A −1 A =
−3 2 3 5

E
Use a calculator if you
5 × 2 + (−3) × 3 5 × 3 + (−3) × 5 1 0
wish. = =
(−3) × 2 + 2 × 3 (−3) × 3 + 2 × 5 0 1
∴ A −1 A = I
Like matrix multiplication, finding the inverse of a matrix generally requires long and tedious
If A =
a b
PL
computations. Fortunately, a graphics calculator will do it for us. However, not all square
matrices have inverses. To see why, we need to introduce another new matrix concept, the
determinant of a square matrix and see how it is related to finding the inverse of a square
matrix. To keep things manageable, we will restrict ourselves initially to 2 × 2 matrices.
The determinant of a matrix
of a 2 × 2 matrix
The determinant

, then the determinant of matrix A, written as det (A) or |A|, is given by
M
c d

a b

det (A) = =a×d −b×c
c d
Example 3 Finding the determinant of a 2 × 2 matrix

SA
Find thedeterminant
of the matrices:

2 3 2 3 2 4
a A= b B= c C=
3 5 2 3 2 3
Solution

1 Write down the matrix 2 3
2 3
a A= ∴ det ( A ) = =2×5−3×3=1
and use the rule 3 5 3 5
a b

det (A) = 2 3
c d 2 3
b B = ∴ det ( B ) = =2×3−2×3=0
= a × d − b × c. 2 3 2 3

2 Evaluate. 2 4
2 4
c C = ∴ det (C ) = = 2 × 3 − 2 × 4 = −2
2 3 2 3
P1: FXS/ABE P2: FXS
Back to Menu >>>

From Example 3, we can see that the determinant of a matrix is a number that can take on both
positive and negative values as well as being zero. As we will now see, for a matrix to have an
inverse, its determinant must be non-zero.
How to determine the inverse of a matrix
of a2 × 2 matrix
The inverse
a b
If A = , then its inverse A−1 , is given by
c d
E

1 d −b 1 d −b
A−1 = =
ad − bc −c a det (A) −c a
1
provided = 0, that is, provided det (A) = 0.
ad − bc
Example 4
a A=
Solution
2 2
3 4
PL
Finding the inverse of a 2 × 2 matrix
Find the inverse of the matrices:

b B=

2
2
3
3

c C=

2
2
4
3

M
1 Write down the matrix a 2 2
A =
and use the rule 3 4
a b
2 2
det (A) =
c d ∴ det ( A ) = =2×4−3×2=2
3 4
= a × d − b × c.
Evaluate.
SA
Use the formula

−1
1 d −b 1 4 −2
1 d −b ∴A = =
−1
A = det ( A ) −c a 2 −3 2
det (A) −c a
to evaluate. 2 −1
=
−1.5 1

2 Write down the matrix and use the b 2 3
a b B =
2 3
rule det (B) = = a × d − b × c.
c d 2 3

Evaluate. ∴ det ( B ) = =2×3−2×3=0
2 3
As the determinant of the matrix is zero,
Inverse not defined
the matrix has no inverse.
P1: FXS/ABE P2: FXS
Back to Menu >>>


3 Write down the matrix and c 2 4
use the rule C =
2 3
a b
2 4
det (C) =
c d ∴ det (C ) = = 2 × 3 − 2 × 4 = −2
2 3
= a × d − b × c.
Evaluate. −1
1 d −b
∴C =
Use the formula det (C ) −c a

1 d −b 1 3 −4 −1.5
C −1 = =
=
2
det (C) −c a −2 −2 −1
E
2 1
to evaluate.
Using a graphics calculator to determine the determinant and

inverse of an n × n matrix (n ≥ 2)
PL
There are rules for finding the inverse of a square matrix of any size, but they are extremely
complicated and take huge amounts of time to compute by hand. So, in practice, we use a
calculator to find the inverse of all but a (2 × 2) matrix. In fact, you can use a calculator to
compute the inverse of a (2 × 2) matrix if you wish, but it is often just as quick to do it by
hand. The same goes for calculating determinants, although the determinant of a (2 × 2)
matrix is much more quickly computed by hand.
How to find the determinant and inverse of a matrix using the TI-Nspire CAS
M
 
1 2 3
 
If A =  4 1 0 , find det (A) and A−1 .
2 0 2
Steps
SA
1 Start a new document and select 1:Add

Calculator.
2 Enter the matrix A into your calculator.
3 To calculate det(A), type det(a) and
press enter to evaluate.
Note: det( ) can also be accessed using
b /7:Matrix & Vector/2:Determinant.
P1: FXS/ABE P2: FXS
Back to Menu >>>

4 To calculate the inverse matrix A−1 ,

type a ∧ − 1 and press enter to evaluate.
If you want to see the answer in
fractional form, enter as exact(a ∧ − 1)
and press enter to evaluate.
Notes:
1 When the elements in the matrix to be
E
inverted are whole numbers, the elements
of the inverse will always be whole
numbers or fractions. If this is the case,
and the inverse you obtain with your
calculator contains decimals, it is worth
PL
converting it to fractional form.
Although the mode is set to APPRX (or
decimal), we can make the display show
fractions using exact( . . ..) in front of the
command. exact( . . . ) can also be pasted
from the Catalog ( ).
2 If the matrix has no inverse, the
calculator will respond with the error
message Singular matrix.
M
How to find the determinant and inverse of a matrix using the ClassPad
 
1 2 3
SA
 
If A =  4 1 0 , find det (A) and A−1 .
2 0 2
Steps
1 Enter the matrix A into your
calculator.
Note: Change the status of the
calculator to Standard in order for
fractions to be displayed. Tapping on
Decimal will change the calculator to
Standard.
P1: FXS/ABE P2: FXS
Back to Menu >>>

2 To calculate det(A), type A,

highlight the A (by swiping
with the stylus), select
Interactive from the menu
bar, choose the menu entry
of Matrix-Calculation, then
tap det.
E
3 To calculate the inverse
matrix A−1 , type A ∧ − 1.
PL
Press E to evaluate.
Note: If the matrix has no inverse,
the calculator will respond with the
message Undefined.
M
Exercise 27A
1 a Write down the:
SA
i 2 × 2 identity matrix ii 3 × 3 identity matrix iii 4 × 4 identity matrix

1 2
b If A = , show that AI = I A = A
0 3
 
1 2 0
 
c If C =  3 1 0 , show that C I = I C = C
0 1 2
2 Show that each of the following pairs of matrices are inverses by multiplying one by the
other. Use a calculator if you wish.

1 1 2 −1 2 2 2 −1
a and b and
1 2 −1 1 3 4 −1.5 1

9 7 −3 7 0 2 0 0.5
c and d and
4 3 4 −9 2 0 0.5 0
P1: FXS/ABE P2: FXS
Back to Menu >>>

 1 
4 1 1 
− −
1 1  3 10 1  9 45 
and  3
1 and 
2
e f
1 4 1 5 5 1
− −
3 3 9 9
3 Determine (by hand) the value of the determinant for each of the following matrices:

1 2 0 3 1 2 −1 2
a A= b B= c C= d D=
0 3 1 4 2 4 2 4

−1 2 1 0 0 −3 5 2.5
e E= f F= g G= h H=
E
−2 4 0 3 1 4 4 2

1 0 0 −2
i I = j J=
0 1 −2 4
4 Use a calculator to determine the inverse of each of the following

matrices:
a A=



1.1 2.2
0 3.0
1 1 0
d D = 1 1 0
0 1 1



PL
b B=



0.2 −0.1
10
2 1
e E =  0 1 −1 
0 0
4
0
1



c C= 2

1
2
1
2
4

Applications of the inverse matrix: solving

M
27.2
simultaneous linear equations
Another application of inverse matrices is the solution of simultaneous linear equations.
In the previous chapter, we saw that the pair of simultaneous linear equations:
4x + 2y = 5
SA
3x + 2y = 2
could be written as the matrix equation:

4 2 x 5
= .
3 2 y 2
We will now show that inverse matrices can be used to solve this matrix equation.

4 2 x 5
Let A = , X= and C = .
3 2 y 2
Then, AX = C.
Assuming that A−1 , the inverse of A, is defined we have:
A−1 AX = A−1 C
or I X = A−1 C
or X = A−1 C
Cambridge University
Note: The order ofPress • Uncorrected
multiplication Sample here.
is important A−1 C not C A−1 . • 2008 © Jones, Evans, Lipson
pagesIt•is978-0-521-61328-6
P1: FXS/ABE P2: FXS
Back to Menu >>>


−1 1 −1 x 5
Noting A = and substituting X = and C = ,
−1.5 2 y 2

x 1 −1 5
=
y −1.5 2 2

1 × 5 + (−1) × 2 3
= = (or use a calculator)
(−1.5) × 5 + 2 × 2 −3.5

x 3
∴ = or x = 3 and y = −3.5
E
y −3.5
Thus, using the matrices, the process of solving a pair of simultaneous linear equations can be
reduced to finding the inverse of a matrix and performing a matrix multiplication. While there
is not a lot to gain over algebraic methods for a pair of simultaneous equations, the same
PL
matrix method can be used to solve three, four or more simultaneous linear equations. This is
an extremely difficult task using algebraic methods.
Before we go on to practise the method, it is important for you to know when it does not
work and why. First, we need to recognise that not all pairs of simultaneous linear equations
have solutions. This can happen for two reasons:
the equations are inconsistent, their graphs are parallel and thus do not cross
the equations are dependent, the graphs coincide and thus do not cross at a single point
Inconsistent equations y
M
For example, the pair of simultaneous linear equations
2
x + 2y = 4
2x + 4y = 6 1.5
x + 2y = 4
have no solution. We can see why, by plotting their graphs
2x + 4y = 6
(opposite). When plotted, the two equations are parallel
x
SA
lines with no point of intersection. Sets of equations that O 3 4

have this property are said to be inconsistent.
If we try to solve these equations using matrix Calculator hint
methods by solving the matrix equation: To construct this plot on a graphics
calculator, rewrite the equations in
1 2 x 4
= . the form y = mx + c.
2 4 y 6
x + 2y = 4
we find that it cannot be done. This is because the or 2y = 4 − x or y = 2 − 0.5x
determinant of the coefficient matrix 2x + 4y = 6

1 2 or 4y = 6 − 2x or y = 1.5 − 0.5x

det (A) = = 1 × 4 − 2 × 2 = 0.
2 4
As a general rule, if a pair of linear equations is
inconsistent, det (A) = 0 and the equations have
no solution.
P1: FXS/ABE P2: FXS
Back to Menu >>>

Dependent equations
For example, the pair of simultaneous linear equations
y
x + 2y = 4
2x + 4y = 8
2
x + 2y = 4
have no solution. We can see why, by plotting a graph
(opposite). When plotted, the graphs of the two equations
coincide, with no single point of intersection. 2x + 4y = 8
E
There is no unique solution. Rather, there are an x
O 4
infinite number of possible solutions, all the pairs of
points that lie on the line. When this happens, we
say that the equations are dependent.
If we try to solve these equations using matrix Calculator hint

1 2
2 4
PL
methods by solving the matrix equation:

x
y
=
4
8
.
we find that it cannot be done. As before, the

determinant of the coefficient matrix
det(A) =

1 2

2 4

= 1 × 4 − 2 × 2 = 0.
To construct this plot on a graphics
calculator, rewrite the equations in
the form y = mx + c.
x + 2y = 4
or 2y = 4 − x or y = 2 − 0.5x
2x + 4y = 8
or 4y = 8 − 2x or y = 2 − 0.5x
M
As a general rule, if a pair of linear equations is
dependent, det(A) = 0 and the equations have
no unique solution.
SA
Solving a set of linear equations

Provided, det (A) = 0, the set of linear equations defined by the matrix equation AX = C
has the unique solution
X = A−1 C
Note: The order of multiplication is important here: X = A−1 C not CA−1
For the set of equations:
ax + by = e a b x e
A= , X= and C = .
cx + dy = f c d y f
For the set of equations:      
ax + by + cz = j a b c x j
     
d x + ey + f z = k A = d e f  , X =  y  and C =  k ,
gx + hy + i z = l g h i z l
and so on.
P1: FXS/ABE P2: FXS
Back to Menu >>>

Example 5 Solving a set of two simultaneous linear equations using

the inverse matrix
Solve using matrix methods:

3x + 4y = 6
2x + 3y = 4
Solution
3 4 x 6
1 Rewrite the equations in matrix form. =
E
2 3 y 4
2 By identifying the matrices, A, X, and
3 4 x 6
C, rewrite the matrix equation in the Let A = , X = , C =
2 3 y 4
form AX = C.
Then AX = C or X = A −1 C (det ( A ) = 0)
PL
3 Provided det (A) = 0, the solution
X = A−1 C.
4 Enter the matrices A and C into your

calculator.
5 Solve the matrix equation by evaluating
the matrix product A−1 C.
M

6 Write down your answer. 2
Therefore, X = or x = 2 and y = 0.
0
However, not all pairs of simultaneous linear equations have unique solutions, as we will see in
SA
the next example .
Example 6 A set of two simultaneous linear equations with no unique solution
3x + 4.5y = 9
2x + 3y = 4
P1: FXS/ABE P2: FXS
Back to Menu >>>

Solution
1 Rewrite the equations in matrix form.
3 4.5 x 9
=
2 3 y 4
2 By identifying the matrices, A, X,
3 4.5 x 9
and C, rewrite the matrix equation Let A = , X = , C = .
2 3 y 4
in the form AX = C.
3 Provided det (A) = 0, the solution Then AX = C or X = A −1 C (det ( A ) = 0)
in matrix form is X = A−1 C.
E
In this case:

3 4.5
det (A) = = 0,
2 3
but let us see what happens.
your calculator.
PL
4 Enter the matrices A and C into
5 Attempt to solve the matrix equation

by evaluating the matrix product A−1 C.
6 The calculator gives an error
This is because det(A) = 0. The system

of equations does not have a unique
solution. Write down your conclusion.
No unique solution as det(A ) = 0.
message: Singular matrix or Undefined MATrix.
M
The power of the matrix method for solving systems of linear equations becomes apparent
when we solve a system of three or more equations.
Example 7 Solving a set of three simultaneous linear equations using

the inverse matrix
SA
3x + 4y − 2z = −5
2x + 3y = −1
x + 2y + 3z = 3
Solution
    
1 Rewrite the equations in matrix form. 3 4 −2 x −5
    
2 3 0   y  =  −1 
1 2 3 z 3
     
2 By identifying the matrices, 3 4 −2 x −5
     
A, X, and C, rewrite the matrix Let A =  2 3 0  , X =  y  , C =  −1 
equation in the form AX = C. 1 2 3 z 3
Note: There is no z term in the second equation, so its coefficient is zero.
P1: FXS/ABE P2: FXS
Back to Menu >>>

3 Provided det(A) = 0, the solution in Then AX = C or X = A −1 C (det(A ) = 0)

−1
matrix form is X = A C.
4 Enter the matrices A and C into your calculator.
Note: Refer to page 639 if you are unsure how
to do this.
5 Solve the matrix equation by evaluating the
E
matrix product A−1 C.
Example 8
PL
6 Write down your answer.
Application

−11
Therefore, X =  7 
0

or x = −11, y = 7 and + z = 0.
M
A manufacturer makes two sorts of orange flavoured chocolates, House Brand and Orange
Delights. The number of kilograms of House Brand (x) and the number of kilograms of
Orange Delights (y) that can be made from 80 kg of chocolate and 120 kg of orange fill can be
found by solving the following system of equations:
0.3x + 0.5y = 80
SA
0.7x + 0.5y = 120
Solve for x and y using matrix methods.
Solution
1 Rewrite equations in matrix form.

0.3 0.5 x 80
=
0.7 0.5 y 120

2 By identifying the matrices, A, X, 0.3 0.5 x 80
and C, rewrite the matrix equation
Let A = ,X = ,C = .
0.7 0.5 y 120
in the form AX = C.
3 Provided det(A) = 0, the Then AX = C
solution X = A−1 C. or X = A −1 C (det(A ) = 0)
P1: FXS/ABE P2: FXS
Back to Menu >>>

4 Enter the matrices A and C into

your calculator and solve the matrix equation
by evaluating the matrix product A−1 C.

5 Write down your answer. 100
Therefore, X = or x = 100 and y = 100.
100
E
Exercise 27B
a 3x + 2y = 2
g
2x + 5y = 4
x − 3y = 7
−2x + y = 4
5x − 2z = 3
x −y+ z=2
x +y+ z=1
PL
1 Write each of the following systems of linear equations in matrix form.
b 3x + 5y = 6
h
2x + 4y = 3
e −3x − 2y = 2
x + 2y = −1
x
2x
+ y
− y
− 2z
+ z
+ w
− w
=3
=2
c x + 2y = 1
2x − 3y = 2
f 3x + 4y − 2z = 5
2x + 3y + 5z = 2
x + 2y + 3z = 3
x + 2y + z + w =1
M
2x − 3y + 2z − 2w =0
2 Give two explanations of how a system of two linear equations can have no unique solution.
3 What is the condition for the matrix equation AX = C not to have a unique solution?
4 Which of the following matrix equations have no solution?

SA
3 1 x 2 4 2 x 2 −1 1 x 0
a = b = c =
2 1 y 1 2 1 y 1 2 −1 y 1

0 −1 x 2 2.5 1 x 2 1 1 x 0
d = e = f =
2 1 y 4 5 2 y 1 2 2 y 2
    
1 2 3 x 1
−1 1 x 0 5 2 x 10     
g = h = i 2 4 6 y  = 0
2 −1 y 1 2 1 y 4
1 1 1 z 1
         
1 0 3 x 1 −1 0 2 x 2
         
j 0 1 0 y  = 0 k  0 1 1 y  = 1
0 1 1 z 1 0 1 1 z 3
P1: FXS/ABE P2: FXS
Back to Menu >>>

5 Solve the following matrix equations. (Not all have solutions.)

1 1 x 2 4 2 x 2 1 −1 x 2
a = b = c =
2 3 y 1 8 4 y 1 −2 1 y 1

1 −1 x 2 5 1 x 20 3 3 x 6
d = e = f =
2 0 y 4 10 2 y 10 5 5 y 1
     
0 2 1 x 3
−1 1 x 2     
g = h  1 −1 3   y  =  2 
0 −1 y 1
E
1 1 1 z 2
         
1 0 3 x 1 −1 0 2 x 3
         
i 0 1 0 y  = 5 j  0 1 1 y  =  2
2 0 6 z 1 0 1 −1 z −1
6 Solve the following systems of linear equations using matrix methods. (Not all have
solutions.)
a 3x − 5y = 9
e
2x − 3y = 12
PL b x + 3y = 4
x + 5y = 10
c x + 3y = 4
2x + 6y = 5
x + 4y − 2z = 8 f x + y − z = 0 g x + y − z = 2 h −2y + z + w =−5
2x + 3y + 2z = 4
x + 2y − z = 3
x + y+z=2
x + 2y − z = −3 x−
y+z=1
z=2
x + y + z − w =2
x+ y+ 2w =0
x + y + z + w =0
d 3x − 2y = 4
10x − 6y = 9
7 An outdoor clothing manufacturer makes two sorts of jackets, Polarbear and Polarfox. The
M
number of Polarbear jackets (x) and the number of Polarfox jackets (y) that can be made
from 150 metres of fabric and 150 hours of worker time can be found by solving the
following system of equations:
2x + 2y = 150
2.4x + 1.8y = 150
SA
Solve for x and y using matrix methods.
8 An advertising company plans to run an advertising campaign in three regional cities.

Advertisements will be run on TV, radio and in the local newspaper. The costs of running the
same advertising program in the three cities are $175 000, $149 000 and $183 500
respectively. Find the number of advertisements that were run on TV (x), Radio (y) and the
local newspaper (z) by solving the following system of equations:
3000x + 1000y + 500z = 175 000

2500x + 900y + 600z = 149 000
3100x + 1100y + 650z = 183 500
Solve for x, y and z using matrix methods.
P1: FXS/ABE P2: FXS
Back to Menu >>>

27.3 Matrix powers

Power of a matrix
Before we introduce the transition matrix, we need to introduce the idea of the power of a
matrix, An , where n is a positive integer (n = 1, 2, 3, . . .)
Just as we define
22 as 2 × 2,
23 as 2 × 2 × 2,
24 as 2 × 2 × 2 × 2 and so on,
E
we define
A2 as A × A,
A3 as A × A × A,
A4 as A × A × A × A and so on.
PL
Only square matrices can be raised to a power.
Finally, just as we define 20 to be 1, A0 is defined to be I, the identity matrix.
Determining the power of a matrix A

If A is a square matrix, then
A2 = A × A
A3 = A × A × A and so on.
M
A0 is defined as I, the appropriate identity matrix.
Example 9 Determining the power of a matrix

2 3
If A = determine A3 .
4 3
SA
Solution
2 3
1 Write down the matrix. A =
4 3
2 Either:
Evaluate A3 by multiplying out. A3 =
A × A × A
Write down your answer. 92 93
∴ A3 =
124 123
or:
Enter the matrix into your calculator. A ∧3
A3 =
Evaluate A3 by raising to a 92 93
power. Write down your answer. ∴A =
3
124 123
P1: FXS/ABE P2: FXS
Back to Menu >>>

Example 10 Evaluating matrix expressions involving powers

1 0 −1 1 0 1
If A = ,B = and C = determine:
2 −1 2 1 1 1
a 2A + B 2 − 2C b (2A − B)2 − C 2 c AB 2 − 3C 2
Solution
1 0 −1 1 0 1
1 Write down the matrices A = , B = C =
2 −1 2 1 1 1
E
2 Enter the matrices A, B and C
into your calculator.
Write down your answer.

PL
3 Type in each of the expressions
as written and press to evaluate.
Note: For CAS calculators you must

use a multiplication sign between
a and b2 in the last example otherwise
M
it will be read as variable (ab)2 .
5 −2
a 2A + B − 2C = 2
2 −1

6 −1
b (2A − B ) − C =
2 2
−1 5

0 −3
SA
c AB2 − 3C 2 =
3 −9
Exercise 27C

2 1
1 If A = determine A2 , A3 , A4 and A7 .
1 3

−1 1
2 If A = determine A4 , A5 , A6 and A7 .
1 2

2 1 −1 1 0 1
3 If A = , B= and C = , evaluate
1 3 1 2 1 −2
a A + 2B − C 2 b AB − 2C 2 c (A + B + 2C)2
d 4A + 3B 2 − C 3 e (A − B)3 − C 3
P1: FXS/ABE P2: FXS
Back to Menu >>>

27.4 Transition matrices and their applications

Setting up a transition matrix
A car rental firm has two branches, one in Bendigo and one in Colac. Cars are usually rented
and returned in the same town. However, a small percentage of cars rented in Bendigo each
week are returned in Colac, and vice versa. The diagram below describes what happens on a
weekly basis.
B = Bendigo
20%
E
C = Colac
80% B 90%
C
10%
From week to week:

PL
What does this diagram tell us?
0.8 (or 80%) of cars rented each week in Bendigo are returned to Bendigo
0.2 (or 20%) of cars rented each week in Bendigo are returned to Colac
0.1 (or 10%) of cars rented each week in Colac are returned to Bendigo
0.9 (or 90%) of cars rented each week in Colac are returned to Colac
The percentages (written as proportions) are summarised in the form of the matrix below.
rented in
M
Bendigo
Colac

Bendigo 0.8 0.1
returned to
Colac 0.2 0.9
This matrix is an example of a transition matrix. It describes the way in which transitions are
made between two states:
SA
State 1: the rental car is based in Bendigo.

State 2: the rental car is based in Colac.
Note: The columns in a transitional matrix will always add to one (100%). This is because all possibilities must
be taken into account. In this case, if 80% of cars are returned to Bendigo then 20% must be returned to Colac.
Example 11 Setting up a transition matrix
The diagram gives the weekly return rates of 80%

rental cars at three locations, Albury, Wodonga A = Albury
and Benalla. Construct a transition matrix that B = Benalla
10% W 15% W = Wodonga
describes the week-by-week return rates at each 70%
of the three locations. Convert the percentages 5% 11%
A B
to proportions. 20%
12% 77%
P1: FXS/ABE P2: FXS
Back to Menu >>>

Solution rented in
1 There are three locations from which the cars can A W B
 
be rented and returned, Albury (A), Wodonga (W) A
 
and Benalla (B). To take into account all the returned to W  
possibilities, a 3 × 3 transition matrix is needed. B
Constuct a blank matrix labelling the rows and
columns A, B and C respectively. Column labels
indicate where the car was rented. The row
E
labels indicate where the cars were returned to.
2 Complete the matrix by writing each of the 
A W B

percentages (converted to proportions) into the A 0.7
 
appropriate locations in the matrix. Start with W  0.1 
Column A and write in values for each row: B 0.2
PL
0.7 (70%), 0.1 (10%) and 0.2 (20%).
3 Mentally check your answer by summing columns;
they should sum to one.
Example 12 Setting up a transition matrix


A W B
A 0.7 0.05 0.12



W  0.1 0.8 0.11 
B 0.2 0.15 0.77
A factory has a large number of machines. Machines can be in one of two states, operating or
M
broken. Broken machines are repaired and come back into operation and vice versa. On a
given day:
85% of machines that are operational stay operating
15% of machines that are operating break down
5% of machines that are broken are repaired and start operating
95% of machines that are broken stay broken.
SA
Construct a transition matrix to describe this situation. Use the columns to define the situation
at the ‘Start’ of the day and the rows to describe the situation at the ‘End’ of the day.
start
Solution
O B
1 There are two machine states, operating (O) or broken (B). O
To take into account all the possibilities, a 2 × 2 transition end
B
matrix is needed. Constuct a blank matrix, labelling the rows
and columns O and B respectively.
2 Complete the matrix by writing each of the percentages O B
O 0.85
(converted to proportions) into the appropriate locations
B 0.15
in the matrix. Start with Column O and write in values for
each row: 0.85 (85%) and 0.15 (15%).
3 Mentally check your answer by summing columns; they O B
O 0.85 0.05
should sum to one.
B 0.15 0.95
P1: FXS/ABE P2: FXS
Back to Menu >>>

Using a transition matrix to make predictions

Let us return to the rental car problem at the start of this section. A car rental firm has two
branches, one in Bendigo and one in Colac. Cars are usually rented and returned in the same
town. However, a small percentage of cars rented in Bendigo each week are returned in Colac,
and vice versa. The diagram below describes what happens on a weekly basis.
B = Bendigo
20%
C = Colac
E
80% B 90%
C
10%
The car rental firm now plans to buy 90 new cars and base 50 in Bendigo and 40 in Colac.
PL
Given this pattern of rental car returns, the questions the manager would like answered are:
‘If we start off with 50 cars in Bendigo, and 40 cars in Colac, how many cars will be
available for rent at Bendigo and Colac after one week, two weeks, etc?’
‘What will happen in the long term? Will the numbers of cars available for rent each week
from each location vary from week to week or will it settle down to some fixed value?’
Using our knowledge of matrices, we can answer both these questions.
The transition matrix that describes how the cars move between Bendigo and Colac is

0.8 0.1

T =
M
0.2 0.9
We start off with 50 cars in Bendigo, and 40 cars in Colac.

Construct a column matrix S0 , called an initial state matrix, to show this situation.

50
S0 =
40
SA
After one week, the number of cars at each branch will change.
Construct a new state matrix S1 to show the number of cars at each branch after one week.
Then,
S1 = T S0

0.8 0.1 50
=
0.2 0.9 40

0.8 × 50 + 0.1 × 40
=
0.2 × 50 + 0.9 × 40

44
or S1 =
46
Thus, after one week there will be 44 cars in Bendigo and 46 in Colac.
P1: FXS/ABE P2: FXS
Back to Menu >>>

What is the situation after two weeks?

Following the same pattern,

0.8 0.1 44 39.8
S2 = T S1 = =
0.2 0.9 46 50.2
Thus, after two weeks (theoretically) there will be 39.8 cars in Bendigo and 50.2 in Colac.
After three weeks

0.8 0.1 39.8 36.9
S3 = T S2 = =
E
0.2 0.9 50.2 53.1
Thus, after three weeks (theoretically) there will be 36.9 cars in Bendigo and 53.1 in Colac.
If we keep on with the process we find the following:
Week 4 Week 5 Week 6 Week 7 Week 11

Week 12
Week 13
Week 14 Week 15

34.8
55.2
33.4
56.6
32.4
57.6
31.6
58.4
PL
...
30.4
59.6
30.3
59.7
30.2
59.8
30.1
59.9
30.1
59.9
What you should notice is that as the weeks go by, the number of cars at each of the
starts to settle down to what we call the steady state solution given by the matrix
locations
30.1
59.9

The theoretical steady state solution is 30.1 (in practice 30) cars at the Bendigo branch and
59.9 (in practice 60) cars at the Colac branch and, it will not change from then on. This can
be seen in the graph (the points have been joined to guide the eye).
.
M
Initial value Bendigo: 50 Steady state value: Colac 60
60
Colac
50
Number of cars
40
SA
Bendigo
30
Initial value Colac: 40 Steady state value: Bendigo 30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Weeks
P1: FXS/ABE P2: FXS
Back to Menu >>>

Example 13 Making predictions with a transition matrix
Let us return to the factory problem in Example 12. The factory has a large number of
machines. The machines can be in one of two states, operating (O), or broken (B). Broken
machines are repaired and come back into operation and vice versa. On a given day, the
situation is described by the transition matrix. Columns define the machine states at the start of
the day.
O B
E
O 0.85 0.05
T =
B 0.15 0.95
At the start of a particular day, 80 machines are operating and 20 are broken. How many
machines are in operation and how many are broken after:
a one day?
Solution
PL
b three days?
1 Write down the transition matrix.
2 Write down a column matrix with

S0 representing the initial
operational state of the machines.
T =
S0 =

0.85 0.05
0.15 0.95
80
20

M

3 To determine the operational state 0.85 0.05 80 69
of the machines after one day, form S1 = T S0 = =
0.15 0.95 20 31
the product S1 = T S0 and evaluate.
4 Write down your conclusion. After one day, 69 machines are operational
and 31 are broken.

SA
5 To find the operational state of 0.85 0.05 69 60.2

the machines after three days, S2 = T S1 = =
0.15 0.95 31 39.8
we must first find the operating state
of the machines after two days.
Form the product S2 = T S1 and
evaluate.

6 The operating state of the 0.85 0.05 60.2 53.16
machines after 3 days is then S3 = T S2 = =
0.15 0.95 39.8 46.84
given by the product S3 = T S2 .
Evaluate.
7 Write down your conclusion. After three days, 53 machines are
operating and 47 are broken.
P1: FXS/ABE P2: FXS
Back to Menu >>>

While we can use repeated matrix multiplication to work out successive states of an evolving
situation such as the car rental problem, it is clearly extremely tedious. Fortunately there is a
more efficient way of doing this and working out what happens in the long term.
If we follow through the process step by step we have:
S1 = T S0
S2 = T S1 = T (T S0 ) = T 2 S0
S3 = T S2 = T (T S1 ) = T 2 S1 = T 2 (T S0 ) = T 3 S0 and so on.
Thus we can write for the nth step:
E
Sn = T n S0
We now have a simple rule for determining the state matrix after n steps.
Example 14
PL Making predictions with a transition matrix
Let us return to the factory problem of Examples 12 and 13.

On a given day, the situation is described by the transition
matrix. Columns define the machine states at the start of a day.
O
T = O 0.85
B 0.15
At the start of a particular day, 80 machines are operating and 20 are broken. How many
machines are in operation and how many are broken after 10 days?
Solution
B
0.05
0.95
M
1 Write down the transition matrix T and initial
0.85 0.05 80
state matrix S0 . Enter the matrices into your T = S0 =
0.15 0.95 20
calculator. Use T and S.
2 To find out how many machines are in Sn = T n S0
operation and how many are broken after
10 days, write down the rule Sn = T n S0 and ∴ S10 = T 10 S0
SA
substitute n = 10 to give S10 = T 10 S0 .
3 Enter the expression T 10 S into your calculator

and evaluate.
Hint: Use the mode key to set your calculator to give
answers to one decimal place.
4 Write down your answer in matrix form,

30.9
then in words. S10 =
69.1
After 10 days, 31 machines will be
operational and 69 broken.
P1: FXS/ABE P2: FXS
Back to Menu >>>

Finding the steady state solution

Initial value Bendigo: 50 Steady state value: Colac 60
In the car rental problem, we found that even 60
Colac
though the number of cars returned to each 50
Number of cars
40
location varied from day to day, eventually Bendigo
30
it settled down to a steady state solution in 20
Initial value Colac: 40 Steady state value: Bendigo 30
which the number of cars at each location 10

0
remained the same. See the graph opposite. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Weeks
Although we arrived at this conclusion by repeated calculations, we can arrive at the solution
E
much faster.
Finding the steady state solution

If S0 is the initial state matrix, then the steady state matrix S, is given by
S = T n S0
PL
as n tends to infinity (∞).
While in practice we cannot evaluate T n for n = ∞, we find that in many situations, values
of n around 15 to 30 will give a very close approximation to the steady state solution.
Example 15 Estimating the steady state solution
For the car rental problem the transition matrix is T =

0.8 0.1

and S0 =

50

.
M
0.2 0.9 40
Estimate the steady state solution by calculating Sn for n = 10, 15 and 20.
Solution
0.8 0.1 50
1 Write down the transition matrix T and T = S0 =
0.2 0.9 40
initial state matrix S0 . Enter the matrices
SA
into your calculator. Use T and S.

2 Use the rule Sn = T n S0 to write down the Sn = T n S0
expression for the nth state for n = 10. ∴ S10 = T 10 S0
3 Enter the expression T 10 S into your calculator
and evaluate.
Hint: Use the mode key to set your calculator
to give answers to one decimal place.
4 Repeat the process for n = 15 and 20.
P1: FXS/ABE P2: FXS
Back to Menu >>>


5 Write down your answer in matrix 30.6 30.1 30.0
form, then in words. This result S10 = , S15 = , S20 =
59.4 59.9 60.0
agrees with the graphical result
There appears to be a steady state solution
arrived at earlier.
with 30 cars at Bendigo and 60 at Colac.
Exercise 27D
E
1 Complete the following transition matrices:

0.75 0.05 0.90 0.15 0.80 0.50 0.33
a b c d
0.25 0.85 0.20 0.65
2 The diagrams below describe a series of transitions between the states indicated. Construct a
c
40% A
PL
transition matrix that can be used to represent each of these diagrams. Use columns to define
the starting points. Convert the percentages to proportions.
a 60%
55%
70%
B 45%
b
d
70% X
30%
25%
45%
Y 75%
M
10% Y 25% B
15% 45% 20%
60%
15% 23% 35% 20%
X Z A C
30% 30%
55% 65%
22% 15%
3 Two fast-food outlets, Jill’s and Pete’s, are located in a small town. In a given week:
SA
r 80% of people who go to Jill’s return the next week

r 20% of people who go to Jill’s go to Pete’s the next week
r 25% of people who go to Pete’s go to Jill’s the next week
r 75% of people who go to Pete’s return the next week
a Construct a transition matrix T of the form

This week
J P

J
Next week:
P
to describe this situation.

b In the initial week, 800 people go to the fast-food outlets, 400 to Jill’s and 400 to Pete’s.
Write down a column matrix S0 that describes this situation.
P1: FXS/ABE P2: FXS
Back to Menu >>>

c How many of the 800 people do we expect to go to Jill’s in the next week? How many go
to Pete’s?
d How many of the 800 people do we expect to go to Jill’s after five weeks’ time? How
many go to Pete’s?
e In the longer term, how many of the 800 people do we expect to go to Jill’s? How many
go to Pete’s?
4 Imagine we live in a world in which people are either ‘happy’ or ‘unhappy’. However, the
way people feel changes from day to day. In the imagined world:
E
r 90% of people who are happy today will be happy tomorrow
r 10% of people who are happy today will be unhappy tomorrow
r 40% of people who are unhappy today will be happy tomorrow
r 60% of people who are unhappy today will be unhappy tomorrow
b
Tomorrow:
PL
a Construct a transition matrix T of the form
H
U
Today
H

U
On a given day, out of 2000 people, 1500 are happy and 500 are unhappy. Write down a
column matrix S0 that describes this situation.
M
c The next day, how many people do we expect to be ‘happy’ and how many ‘unhappy’?
d After four days, how many people do we expect to be ‘happy’ and how many ‘unhappy’?
e In the long term, how many people do we expect to be ‘happy’ and how many ‘unhappy’?
5 In another model of this world, people can be ‘happy’, ‘neither happy nor sad’, or ‘sad’.
However, the way people feel changes from day to day. In the imagined world:
r 80% of people who are happy today will be happy tomorrow
SA
r 15% of people who are happy today will be neither happy nor sad tomorrow
r 5% of people who are happy today will be sad tomorrow
r 40% of people who are neither happy nor sad today will be happy tomorrow
r 30% of people who are neither happy nor sad today will be the same tomorrow
r 30% of people who are neither happy nor sad today will be sad tomorrow
r 35% of people who are sad today will be happy tomorrow
r 40% of people who are sad today will be neither happy nor sad tomorrow
r 25% of people who are sad today will be sad tomorrow
a Construct a transition matrix T to describe this situation. Use the columns to define the
situation today and the rows to describe the situation tomorrow.
P1: FXS/ABE P2: FXS
Back to Menu >>>

b On a given day, out of 2000 people, 1200 are ‘happy’, 600 are ‘neither happy nor sad’ and
200 are ‘sad’. Write down a column matrix S0 that describes this situation.
c The next day, how many people do we expect to be happy, neither happy nor sad, or sad?
d After five days, how many people do we expect to be happy, neither happy nor sad, or sad?
e In the long term, how many of the 2000 people do we expect to be happy, neither happy
nor sad, or sad?

0.9 0.2 100
6 For the transition matrix T = and an initial state matrix S0 = :
0.1 0.8 200
E
a Use the relationship Sn = T Sn−1 to determine: i S1 ii S2 iii S3
b Determine the value of T 5 .
c Use the relationship Sn = T n Sn−1 to determine i S2 ii S3 iii S7
d Calculate Sn = T S0 for n = 10, 15, 20 and 25 to show that the steady state solution is
n

200
close to
100
PL
.


0.7 0.4 0.1


a Use the relationship Sn = T Sn−1 to determine: i S1

b Use the relationship Sn = T n Sn−1 to determine: i S2
ii S2
ii S3
iii S3
iii S7


100
300


7 For the transition matrix T =  0.2 0.1 0.3  and an initial state matrix S0 =  200 :
0.1 0.5 0.6
c Calculate Sn = T S0 for n = 10, 15, 20 and 25 to show that the steady state solution is
n
 
M
247.1
 
close to  129.4 .
223.5
SA
P1: FXS/ABE P2: FXS
Back to Menu >>>

Review
Key ideas and chapter summary
The identity matrix An identity (unit) matrix I is a square matrix with ones
down the leading diagonal and zeros elsewhere.
 
  1 0 0 0
1 0 0
1 0   
 0 1 0 0
 0 1 0    are all examples of
0 1 0 0 1 0
0 0 1 identity matrices
0 0 0 1
E
The determinant of a matrix The determinant of a matrix A is written det(A) or |A|.

a b a b
If A = , then det(A) = =a×d −b×c
c d c d
Only square matrices have determinants.
The inverse of a matrix

PL For example, if A =

1 3

det (A) =
3 4
the determinant.

1 3
3 4
then
= 1 × 4 − 3 × 3 = −5
For higher order matrices, a calculator is used to calculate
The inverse of a matrix A is written as A−1 and has the

property that A A−1 = A−1 A = I.
Only square matrices have inverses.
M
The inverse of a matrix is not defined if det(A) = 0.

a b
Determining the inverse If A = , then its inverse A−1 , is given by
of a matrix c d

1 d −b 1 d −b
A−1 = =
ad − bc −c a det(A) −c a
SA
1
provided = 0, that is, provided det (A) = 0.
ad − bc

1 3
For example, if A = then
3 4

1 d −b 1 4 −3
A−1 = =
det(A) −c a −5 −3 1

−0.8 0.6
=
0.6 −0.2
For higher order matrices, a calculator is used to determine
the inverse.
P1: FXS/ABE P2: FXS
Back to Menu >>>

Review 750 Essential Further Mathematics — Module 6 Matrices and applications
ax + by = e
Representing systems of linear The set of linear equations can be written in
cx + dy = f
equations with matrices
a b x e
matrix form as = .
c d y f
x + 3y = 1
For example, the set of equations
3x + 4y = 2

1 3 x 1
can written in matrix form as =
E
3 4 y 2
A similar pattern follows for sets of three, four, five etc.,
equations.
Matrix solution of a system Provided det(A) = 0, the set of linear equations defined
of linear equations by AX = C has the solution X = A−1 C. The order of
PL multiplication is important.
For example, the solution of
is given by

x
y
=
=

1 3
3 4
−0.8
1 3
3 4
−1
0.6 −0.2
0.6
1
2

x
y

1
2
=
=
1
2
0.4
0.2

If det(A) = 0, then the system of equations has no unique

M
solution. The equations are either inconsistent (at least two
of the graphs do not cross), or dependent (at least two of the
graphs are identical).
Power of a matrix The power of a matrix is defined in the same way as the
powers of numbers: A2 = A × A, A3 = A × A × A,
SA
A4 = A × A × A × A and so on.
Only square matrices can be raised to a power.
A0 is defined to be I, the identity matrix.
Transition matrix Transition matrices are square and have the property that
the sum
of the columns
equals one. For example,
0.7 0.1
T = could be a transition matrix.
0.3 0.9
Initial state matrix The initial state matrix S0 defines the starting state of a

100
system. For a two-state system, S0 = , could be an
200
initial state matrix.
P1: FXS/ABE P2: FXS
Back to Menu >>>

Review
Steady-state solution matrix The steady-state matrix S represents the final state of a
system. The final state of a system can be estimated by
calculating T n S0 , for a large value of n.
Skills check
Having completed this chapter you should be able to:
E
recognise an identity matrix
calculate the determinant of a matrix
know the properties of an inverse matrix
find the inverse of a square matrix using a calculator
use determinants to test a system of linear equations for solutions

PL
use inverse matrices to solve systems of linear equations
determine the power of a square matrix using a calculator
find the steady-state solution of a system given the initial state S0 , and the transition
matrix T
Multiple-choice questions
The following matrices are needed for Questions 1 to 7 



1 0 2
M
2 0 43 0.9 0.5   1 2
U= V = W = X = 3 2 0 Y =
1 1 45 0.1 0.5 2 4
1 1 2
1 The matrix that cannot be raised to a power is:
A U B V C W D X E Y
2 det(U ) =
SA
A 1 B 0 C 1 D 2 E 4
−1
3 Y =
0.5 0 4 −2 1 0 1 −4 2
A B C D
−0.5 1 −2 1 0 1 8 2 −1
E not defined
4 U −1=
0.5 0 4 −2 1 0 1 −4 2
A B C D
−0.5 1 −2 1 0 1 8 2 −1
E not defined
2
5 3U =
8 0 4 0 12 0 24 0
A B C D
7 1 3 1 9 3 21 3
E not defined
P1: FXS/ABE P2: FXS
Back to Menu >>>

6 The matrix that could be a transition matrix is:

A U B V C W D X E Y
7 The matrix that could be a state matrix with two states is:
A U B V C W D X E Y
8
25%
75% A B 95%
E
5%
The transition matrix that can be used to represent the information in the diagram
above is:
From From
A
A To: A 0.75
B 0.05
A
C To: A 0.75
B 0.25
PL From
From
B
0.25
0.95
B
0.95
0.05
A
B To: A 0.75
B 0.25
A
D To: A 0.75
B 0.95
B
0.05
0.95
From
B
0.25
0.05
M
A B
E To: A 0.25 0.05
B 0.75 0.95
9 60%
SA
10% Y 35%
75%
5% 20%
X Z
15%
50%
30%
The transition matrix that can be used to represent the information in the diagram
above is:
From From
X Y Z X Y Z
   
A To: X 0.75 0.05 0.30 B To: X 0.75 0.10 0.15
   
Y  0.10 0.60 0.20  Y  0.60 0.05 0.35 
Z 0.15 0.35 0.50 Z 0.50 0.30 0.20
P1: FXS/ABE P2: FXS
Back to Menu >>>

Review
From From
X Y Z X Y Z
   
C To: X 0.75 0.10 0.15 D To: X 0.75 0.05 0.15
   
Y  0.60 0.05 0.35  Y  0.10 0.60 0.20 
Z 0.50 0.30 0.20 Z 0.15 0.35 0.50
From
X Y Z
 
E
E To: X 0.75 0.05 0.15
 
Y  0.15 0.35 0.50 
Z 0.10 0.60 0.20
10 Which of the following systems of linear equations has a unique solution?
x − 3y = 6 2x + 2y = 6 4x − 3y = 6
C
I

2x + y = 3
A I only
−3 1
2 3
2 0

x
y
y
PL
11 The linear equations

2 2

x
=
=
II
B I and II only
3

6
3
4x + 4y = 3
2x − 3y = 6
2x + y = 3
6
C II only
D

−3 1
2 −3
III
8x − 12y = 3
D I and III only
can be written in matrix form as:

2 2

y
x

x
=
=
3
6
6
E all
2 1 y 3
M

2 3 x 6
E =
2 1 y 3
The following information is needed for Questions 12 to 15

0.6 0.5 100
SA
A system is defined by a transition matrix T = with S0 = .

0.4 0.5 200
12 For this system, S1 =

60 140 160 166 200
A B C D E
200 160 140 144 100
13 For this system, T 2 is:

0.36 0.25 0.56 0.55 0.6 0.5 1.2 1.0
A B C D
0.16 0.25 0.44 0.45 0.4 0.5 0.8 1.0
E not defined
P1: FXS/ABE P2: FXS
Back to Menu >>>

14 For this system, S3 is closest to:

160 166.6 166.7 640 400
A B C D E
140 133.4 133.3 560 800
15 For this system, the steady state solution is closest to:

166.5 166.6 166.7 166.8 166.9
A B C D E
133.5 133.4 133.3 133.2 133.1
E
Extended-response questions
1 We wish to solve the following system of linear equations
3x + 2y = 7
i
2 3
PL
4x − 2y = 0
using matrix methods.
a Write the equations in matrix form.
b The solution is given by X = A−1 C. Write down the matrices A, A−1 , X , and C.
c Solve the equations.
d Use the determinant test to show that the following systems of linear equations do
not have a unique solution.

x
=
6
ii

2 2

x
=
6
M
6 9 y 3 8 8 y 24

1 −3 x 6
iii =
2 −6 y 3
2 We wish to solve the following system of linear equations
x − 2y + z = 0
SA
3x + 2y − z = 4
2x − y + z = 3
using matrix methods.
a Write the equations in matrix form.
b The solution is given by X = A−1 C. Write down the matrices A, A−1 , X , and C.
c Solve the equations.

0.15 0.75 400
3 For the transition matrix T = and initial state matrix S0 = :
0.85 0.25 800
a Use the relationship Sn = T Sn−1 to determine: i S1 ii S2 iii S3
4
b Determine the value of T .
c Use the relationship Sn = T n Sn−1 to determine: i S2 ii S3 iii S6
d Calculate Sn = T n S0 for n = 10, 15, 30 and 40 to show that the steady state

562.5
solution is close to .
637.5
P1: FXS/ABE P2: FXS
Back to Menu >>>

Review
4 Experience at a fitness centre shows that:
r 70% of members who use the centre in a given week will also use the
centre the next week
r 30% of members who use the centre in a given week will not use the
centre the next week
r 40% of members who do not use the centre in a given week will use the
centre in the next week
E
r 60% of members who do not use the centre in a given week will not use
the centre in the next week
a Construct a transition matrix T of the form:
This week
c
d
PL
Next week: Use
Not use

Use Not use

This week, 400 members used the fitness centre and 100 did not use the fitness
centre. Write down a column matrix S0 that describes this situation.
How many members do you expect to use the centre next week?
How many members do you expect to use the centre after five weeks?
e In the long term, how many members do you expect to use the fitness centre each
M
week?
SA

ch27 Matrices and Applications

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ch27 Matrices and Applications

Uploaded by

Copyright:

Available Formats

P1: FXS/ABE P2: FXS

9780521740517c27.xml CUAU031-EVANS September 7, 2008 7:15

Back to Menu >>>

The inverse matrix

The identity matrix

An identity matrix has the property that, for a square matrix A, IA = AI = A.

Back to Menu >>>

Example 1 Verifying that IA = AI = A

The inverse matrix A−1

Example 2 Verifying that AA-1 = A-1 A = I

Back to Menu >>>

The determinant of a matrix

Example 3 Finding the determinant of a 2 × 2 matrix

Back to Menu >>>

How to determine the inverse of a matrix

Find the inverse of the matrices:

Use the formula

Back to Menu >>>

Using a graphics calculator to determine the determinant and

1 Start a new document and select 1:Add

Back to Menu >>>

4 To calculate the inverse matrix A−1 ,

Back to Menu >>>

2 To calculate det(A), type A,

i 2 × 2 identity matrix ii 3 × 3 identity matrix iii 4 × 4 identity matrix

Back to Menu >>>

4 Use a calculator to determine the inverse of each of the following

Applications of the inverse matrix: solving

could be written as the matrix equation:

Back to Menu >>>

lines with no point of intersection. Sets of equations that O 3 4

Back to Menu >>>

we find that it cannot be done. As before, the

Solving a set of linear equations

Back to Menu >>>

Example 5 Solving a set of two simultaneous linear equations using

Solve using matrix methods:

4 Enter the matrices A and C into your

the next example .

Example 6 A set of two simultaneous linear equations with no unique solution

Solve using matrix methods:

Back to Menu >>>

5 Attempt to solve the matrix equation

This is because det(A) = 0. The system

Example 7 Solving a set of three simultaneous linear equations using

Solve using matrix methods:

Back to Menu >>>

3 Provided det(A) = 0, the solution in Then AX = C or X = A −1 C (det(A ) = 0)

5 Solve the matrix equation by evaluating the

0.7x + 0.5y = 120

Solve for x and y using matrix methods.

Back to Menu >>>

4 Enter the matrices A and C into

4 Which of the following matrix equations have no solution?

Back to Menu >>>

5 Solve the following matrix equations. (Not all have solutions.)

Solve for x and y using matrix methods.

8 An advertising company plans to run an advertising campaign in three regional cities.

3000x + 1000y + 500z = 175 000

Solve for x, y and z using matrix methods.

Back to Menu >>>

27.3 Matrix powers

Determining the power of a matrix A

Example 9 Determining the power of a matrix