Professional Documents
Culture Documents
MAT 260 Module 1 PDF
MAT 260 Module 1 PDF
ZACODE
DODE
Mr. P Banda
Mr. D Kambilombilo
Mr. P. Mwewa
Ms. R Mvula
LINEAR ALGEBRA- MAT 260
Contents
About this MODULE 1
Module overview 3
Unit 1 8
Unit 2 79
How much time you will need to invest to complete the module.
Study skills.
Activity icons.
Units.
1
U LINEAR ALGEBRA- MAT 260
Unit outcomes.
New terminology.
A unit summary.
Resources
For those interested in learning more on contents in this module,
we provide you with a list of additional resources at the end of this
module these may be books, articles or web sites.
Your comments
After completing this module we would appreciate if you would
take a few moments to give us your feedback on any aspect of this
module. Your feedback might include comments on:
Module assignments.
Module assessments.
Module duration.
2
U
LINEAR ALGEBRA- MAT 260
Module overview
Module outcomes
Upon completion of module 1 you will be able to:
3
U
LINEAR ALGEBRA- MAT 260
Time frame
How long?
Study skills
4
U
LINEAR ALGEBRA- MAT 260
responsibilities.
Your most significant considerations will be time and space i.e. the time you
dedicate to your learning and the environment in which you engage in that
learning.
http://www.how-to-study.com/
The “How to study” web site is dedicated to study skills resources. You
will find links to study preparation (a list of nine essentials for a good
study place), taking notes, strategies for reading text books, using
reference sources, test anxiety.
http://www.ucc.vt.edu/stdysk/stdyhlp.html
This is the web site of the Virginia Tech, Division of Student Affairs.
You will find links to time scheduling (including a “where does time
go?” link), a study skill checklist, basic concentration techniques,
control of the study environment, note taking, how to read essays for
analysis, memory skills (“remembering”).
http://www.howtostudy.org/resources.php
Another “How to study” web site with useful links to time management,
efficient reading, questioning/listening/observing skills, getting the most
out of doing (“hands-on” learning), memory building, tips for staying
motivated, developing a learning plan.
The above links are our suggestions to start you on your way. At the time of
5
U
LINEAR ALGEBRA- MAT 260
writing these web links were active. If you want to look for more go to
www.google.com and type “self-study basics”, “self-study tips”, “self-study
skills” or similar.
Need help?
2. email: cosetco@zamtel.zm
Assignments
6
U
LINEAR ALGEBRA- MAT 260
Assessments
There will be one Tutor marked test, at the end of this module, and
several self- marked tasks at the end of each unit.
Assessments A two hour test will be administered at the University College
during a residential school as advised
Margin icons
While working through this module you will notice the frequent use of margin icons. These icons
serve to “signpost” a particular piece of text, a new task or change in activity; they have been
included to help you to find your way around this module.
A complete icon set is shown below. We suggest that you familiarize yourself with the icons and
their meaning before starting your study.
7
U
LINEAR ALGEBRA- MAT 260
Unit 1
Linear equations
1.0 Introduction
From module 1 of MAT 110, you were introduced to linear equations in one and two unknowns
such as ax c and ax by c , where a, b and c are real numbers.
In this unit we will explore further the concepts covered in Linear Algebra in introductory
Mathematics, MAT 110 and look at linear equations in general, that is linear equations in several
variables or unknowns. In this unit we will expose you to the most economic way of
manipulating and solving systems of linear equations.
8
U
LINEAR ALGEBRA- MAT 260
n – component row vector, is a solution of the above equation if, on substituting k i for xi in the
equation a1 k1 a 2 k 2 a3 k 3 a n k n b is true.
The set of values is then said to satisfy the equation.
Example 1.1
Determine the solution of the equation
x 2 y 4z w 3
Solution
To find the solution to this equation, we are looking for real numbers (k1 , k 2 , , k n ) such that
a1 k1 a 2 k 2 a n k n b is a true statement.
If there is no ambiguity about the position of the unknowns in the equation, then we denote this
solution by simply the n-tuple (k1 , k 2, , k n ) .
9
U
LINEAR ALGEBRA- MAT 260
To solve this equation we seek to find a pair of numbers (k1 , k 2 ) (since the equation given is
in two unknowns) which satisfies the equation, that is for which ak1 bk 2 c is a true statement
Thus, solutions of this equation can be obtained by assigning arbitrary values to x and solving
for y and vice versa.
For instance, if we substitute x 0 in the equation, we will get
3(0) 2 y 4
2y 4
y2
Hence (0, 2) is a solution.
Similarly when we substitute x 2 in the equation, we obtain
3(2) 2 y 4
2y 4 6
y 10
y5
Hence (2, 5) is a solution.
The table below gives some possible values for x and the corresponding values for y .
x y
-2 5
-1 4.5
0 2
1 0.5
2 -1
Table 1.1
10
U
LINEAR ALGEBRA- MAT 260
Now any solution (k1 , k 2 ) of the linear equation ax by c determines a point in the
Cartesian plane R2. (i.e in two dimension)
If a and b, are not both zero, the solutions of the linear equation correspond precisely to
the point on a straight line (hence the name linear equation)
A linear equation is said to be degenerate if the coefficients of the unknowns are all zero.
There are two cases when this is possible:
Case I
The constant is not zero that is the equation is of the form 0 x1 0 2 x2 0 n x n b
with b 0
Then there is no solution to the linear equation.
Case II
The constant is also zero that is the equation is of the form
0 x1 0 x 2 0 x n 0
11
U
LINEAR ALGEBRA- MAT 260
a 21 x1 a 22 x 2 a 2 n x n b2
…………………………….....
a m1 x1 a m 2 x 2 a mn x n bm ,
where the aij and b are real numbers. An n – tuple of numbers (k1 , k 2 , , k n ) which satisfies
a 2 x b2 y c2
A pair of numbers which satisfies both equations is called a simultaneous solution of the given
equations or a solution of the system of equations.
There are three (3) cases which can be described geometrically.
(Here we assume that the coefficient of x and y in each equation are not both zero)
Case I.
Consider the equations
x–y=–3
x + 2y = 3
12
U
LINEAR ALGEBRA- MAT 260
Case II
Consider the pair of equations:
x+y=1
2x + 2y = 6
Find the solutions of the given system of equations.
To answer this question we will use the same approach used in the previous question. If we do
the correct thing, we will discover that the system of equation above has no solution.
13
U
LINEAR ALGEBRA- MAT 260
Here the lines corresponding to the linear equations are parallel as shown in figure 1.3 above.
Therefore, the system has no solutions.
Case III
Consider the pair of equations
x y 1
3x 3 y 3
Find the solutions to the system of equation above.
If we look at the system of equations above closely, we will discover that the two equations
represent just one equation, x y 1 .
Refer to figure 1.4 below
14
U
LINEAR ALGEBRA- MAT 260
Cases II and III can occur if and only if (iff) the coefficients of x and the coefficients of
y are proportional:
a1 b
1 .
a 2 b2
There are two possibilities for this to occur:
(i) If the constant terms of the equations are in the same proportion that is
a1 b c
1 1 ,
a 2 b2 c 2
then the lines are coincident.
Hence the system has an infinite number of solutions which correspond to the solutions of either
equation.
15
U
LINEAR ALGEBRA- MAT 260
(ii) If the constant terms of the equations are not in the same proportion, that is:
a1 b c
1 1 ,
a 2 b2 c 2
then the lines are parallel.
Hence the system has no solution.
Example 1.2
a) Determine the solution of the system of equation below:
3x + 6y = 9
2x + 4y = 6
Solution
To find the answer to this problem, we shall compare the ratios of the constants as given above,
that is
a1 b c
1 1
a 2 b2 c 2
3 6 9
2 4 6
3 3 3
2 2 2
Since the ratio of the constants is in the same proportional, then the lines are coincident and the
system has an infinite number of solutions which correspond to the solutions of either equation
by (i) above.
16
U
LINEAR ALGEBRA- MAT 260
Solution
To answer this question we will again compare the ratios of constants as follows:
a1 b c
1 1
a 2 b2 c 2
3 6 9
2 4 5
3 3 9
,
2 2 5
Since the ratios of the constants are not in the same proportional, then the lines are parallel and
the system has no solution by (ii) above.
1 3
2 1
Now since the ratios of the constants are not equal, then the system of linear equations intersect
at one point and as such the system of linear equation has a unique or one solution.
17
U
LINEAR ALGEBRA- MAT 260
Activity 1.1
1. Find the conditions which h and k must satisfy for the following systems of linear
equations to have:
a) A unique solution b) no solution c) an infinite number of solutions.
i) 2 x1 3 x2 x3 5 ii) x1 x 2 4 x3 5 iii) x1 x 2 x3 1
x1 7 x 2 6 x3 h 4 x1 7 x 2 kx3 h x1 hx 2 3 x3 2
2. Determine whether the following sets of equations are
a) Consistent b) inconsistent
i) x1 x 2 2 x3 1 ii) 2 x1 x 2 4 x3 4
2 x1 x 2 x3 2 x1 2 x 2 3 x3 1
iii) x1 2 x 2 x3 2
2 x1 x 2 7 x3 6
x1 x 2 4 x3 4
x1 2 x 2 5 x3 6
18
U
LINEAR ALGEBRA- MAT 260
i) If one of the constants bij 0 , that is the system has an equation of the form:
0 x1 0 x 2 0 x n bi ,
with bi 0 then this equation, and hence the system, has no solution
ii) If every constant bi 0 , that is every equation in the system is of the form:
0 x1 0 x2 0 xn 0
Then this equation, and hence the system, has every n – tuple of real numbers as a solution.
Example 1.3
Determine the solution of the system of equation
0 x 0 y 0 z 0w 0
0 x 0 y 0 z 0w 0
0 x 0 y 0 z 0w 3
Solution
The system has no solution since one of the constants on the right is not zero.
Example
Determine the solution of the system of equation below
0 x 0 y 0 z 0w 0
0 x 0 y 0 z 0w 0
0 x 0 y 0 z 0w 0
Solution
The system has every 4-tuple (k1 , k 2 , k 3 , k 4 ) as a solution since all the constants on the
right are zero.
In the usual case where the equations are not all degenerate, that is, where one of
the aij 0 , we reduce the system to a simpler system.
19
U
LINEAR ALGEBRA- MAT 260
Example 1.4
Find the solution to the system of equation
2x 6 y z 4
3x 2 y z 1
5 x 9 y 2 z 12
Solution
We first observe that all the equations in the system are not degenerate, that is all the aij 0 .
Thus to find the solution, we follow the steps given below:
1
Step 1: Multiply the first equation by so that the system is replaced by
2
1
x 3y z2
2
3x 2 y z 1
5 x 9 y 2 z 12
Step 2: Multiply the new first equation by -3 and add it to the second equation so that the
system is replaced by
1
x 3y z2
2
1
11y z 5
2
5 x 9 y z 12
Step 3: Multiply the first equation by-5 and add it to the third equation so that the system is
replaced by the following which is in the desired form:
1
x 3y z2
2
1
| 11y z 5
2
1
6y z2
2
20
U
LINEAR ALGEBRA- MAT 260
Step 4: Subtract the third equation from the second equation and solving for y as follows:
5 y 7
y 1.4
Step 5: Replace the value of y in the second equation and solving for z, we obtain:
1
11y z 5
2
7 1
11 z 5
5 2
1 77
z 5
2 5
52
z 2
5
104
z
5
z 20.8
Step 6: Finally replace the values of y and z in equation one and solve for x, as follows:
1
x 3y z2
2
1
x 31.4 20.8 2
2
x 4.2 10.4 2
x 12.4 4.2
x 8.2
Therefore the solution to the given system of linear equation is (8.2, 1.4, 20.8)
This process of finding the solution to the system is known as the Gauss Elimination (Reduction)
Method.
The Gauss Elimination method can be summarized by using the following steps:
21
U
LINEAR ALGEBRA- MAT 260
ii) Multiply the first equation by the appropriate non-zero constant so that a11 1
iii) For each i 1 , multiply the first equation by a11 and add it to the i th equation so that the first
unknown is eliminated. Then the system (1) is replaced by an equivalent system of the form:
a 22 x 2 a 23 x3 a 2 n x n b2
-------------------------------------------
a m 2 x 2 a m 3 x3 a mn x n bm
Activity 1.2
Solve the following systems of linear equations using the Gauss elimination method.
i) 2 x y 3z 5 ii) 2x 3 y 2z 5
3x 2 y 2 z 5 x 2 y 3z 2
5 x 3 y z 16 4x y 4z 1
iii) x 2 y z 3w 3
2 x 4 y 4 z 3w 9
3x 6 y z 8w 10
iv) x1 x2 2 x3 3 v) x1 2 x 2 4 x3 7
3x1 4 x 2 5 x3 9 x1 2 x3 2
x1 x 2 x3 6 2 x1 3 x 2 7 x3 9
22
U
LINEAR ALGEBRA- MAT 260
Theorem 1.1
Every system of m equations in n unknowns can be reduced to an equivalent system of the following
form;
x 2 c 23 x3 c 2 r x r c 2 n x n d 2
x3 c3 r x r c3 n x n d 3
x r c rn x n d r
0 d r 1
-----------------------------------------------------------
0 dm
rm
A system of linear equations in the above form, where the leading coefficient is not zero,
is said to be in echelon form. The solutions of a system of equations in the echelon form
can easily be described and found.
If the numbers d r 1 ,, d m are all zero, then the system is said to be consistent and there does
Case I
23
U
LINEAR ALGEBRA- MAT 260
r n , that is, there are as many non-zero equations as unknowns. Then we can successfully
solve uniquely for the unknowns x1 , x r 1 , , x n and so there exists a unique solution for the
system
Case II
r n , that is, there are more unknowns than are non-zero equations. Then we can arbitrary
assign values to the unknowns x r 1 , , x n and then solve uniquely for the unknowns x r , , x1 .
If the numbers d r 1 ,, d m are not all zero that is there is an equation of the form
Example 1.5
a) x 2 y 3z 4 b) x 2 y 3z w 4
0x y 2z 5 0 x y 2 z 3w 5
0x 0 y z 1 0 x 0 y 0 z 0w 0
0x 0 y 0z 0
c) x 2 y 3 z 4
0 x y 2 z 5
0x 0 y 0z 3
Solutions
24
U
LINEAR ALGEBRA- MAT 260
y 2(1) 5
y 3.
x 2(3) 3(1) 4
x63 4
x 1
Thus, x 1 , y 3 and z 1 or in other words, the ordered triple (1, 3,1) is the unique
solution of the system.
b) Since the system has an equation of the form 0 c with c 0 , the system is inconsistent and
has no solution
c) The system is consistent, and since there are more unknowns than non –zero equations the
system has an infinite number of solutions. In fact we can arbitrary assign values to z and w
and solve for x and y.
y 2(1) 3(2) 5
y26 5
y 3
x 2(3) 3(1) 2 4
x 63 2 4
25
U
LINEAR ALGEBRA- MAT 260
x 11
To obtain the general solution to the above system, substitute, say, z a and w b in the
second equation to obtain
y 2a 3b 5
y 5 2a 3b
x 2(5 2a 3b) 3a b 4
x 7a 5b 6
Hence, the general solution of the system is (7a 5b 6, 5 2a 3b, a, b) , where a and b
are real numbers.
Frequently, the general solution is left in terms of z and w (instead of a and b) as follows:
(7 z 5w 6, 5 2 z 3w, z , w)
Activity 1.3
Show that the following sets equations have infinitely many solutions and express the
solutions in terms of parameters.
(i) x1 3x 2 2 ii) x1 x 2 x3 5
2 x1 6 x 2 4 3 x 2 x3 x 4 8
3x 2 x 4 7
26
U
LINEAR ALGEBRA- MAT 260
Theorem 1.2
A system of linear equations belongs to exactly one of the following cases:
Recall that, in the special case of two equations in two unknowns, the above three cases
correspond to the following cases described geometrically:
1.5.1 Definition
a 21 x1 a 22 x 2 a 2 n x n 0
------------------------------------
a m1 x1 a m 2 x2 a mn x n 0
That is if all the constants on the right are zero. By theorem 1.1, the above system is equivalent
to a system of the form:
x2 c2r xr c2n xn 0
27
U
LINEAR ALGEBRA- MAT 260
-------------------------------------------
x r crn x n 0
00
---------------------------------------------
00
Clearly the above system is consistent since the constants are all zero; in fact, the system of
linear equations above, always has the zero solution (0, 0, 0, 0, , 0) which is called the trivial
solution. It has a non – trivial solution if
r n . Hence if we originally begin with fewer equations than unknowns, then r n and so the
system always has a non-trivial solution.
Theorem 1.3
A homogeneous system of linear equations with more unknowns than equations has a non- zero
solution.
Example 1.6
a)
x 2 y 3z w 0
x 3 y z 2w 0
2 x y 3 z 5w 0
b)
x yz 0
2x 3 y z 0
28
U
LINEAR ALGEBRA- MAT 260
x 4 y 2z 0
c)
x yz 0
2x 4 y z 0
3x 2 y 2 z 0
Solutions
a) The homogeneous system has a non – trivial solution since there are more unknowns than
equations.
b) Here the number of non-zero equations is equal to the number of unknowns. Thus, we try
to reduce the given equations in echelon form as follows:
x yz 0
2x 3 y z 0
x 4 y 2z 0
Multiplying the first equation by -2 and adding it to the second equation we obtain;
x yz 0
5 y 3z 0
x 4 y 2z 0
Multiplying the first equation by -1 and adding it to the third equation, we obtain
x yz 0
5 y 3z 0
5 y 3z 0
x yz 0
29
U
LINEAR ALGEBRA- MAT 260
3
y z 0
5
00
From here we conclude that the system has a non – trivial solution since we finally have
only two non –zero equations in three unknowns. For example, choose z 10 , then
solving for y in the second equation. We find
3
y z 0
5
3
y (10) 0
5
y 6
Now replacing y 6 and z 10 in the first equation and solving for x in the first
equation, we obtain
x yz 0
x 6 10 0
x 4
c) Here the number of non-zero equations is equal to the number of unknowns. Thus, we try
to reduce the given equations in echelon form as follows:
x yz 0
2x 4 y z 0
30
U
LINEAR ALGEBRA- MAT 260
3x 2 y 2 z 0
We multiply the first equation by -2 and add it to the second equation. We obtain
x yz 0
2y z 0
3x 2 y 2 z 0
Next we multiply the first equation by -3 and add it to the third equation.
We obtain
x yz 0
2y z 0
y 5z 0
Finally we have
x yz 0
1
y z0
2
y 5z 0
Since in echelon form there are three non-zero equations in three unknowns, the system
has a unique solution, the zero solution (0, 0, 0) .
1.6.1 Definition
Two systems of equations involving the same variables are said to be equivalent if and only if
they have the same solution set.
31
U
LINEAR ALGEBRA- MAT 260
Example 1.7
x1 2 x 2 4
3x1 x 2 2
4 x1 x 2 6
and
4 x1 x 2 6
3x1 x 2 2
x1 2 x 2 4
both involve the same three equations and, consequently, they have the same solution set.
If one equation of a system is multiplied through by a non zero real number, this will have no
effect on the solution set, and the new system will be equivalent to the original system.
Example 1.8
x1 x 2 x3 3
2 x1 x2 4 x3 1
and
2 x1 2 x 2 2 x3 6
2 x1 x2 4 x3 1
If a multiple of one equation is added to another equation, the system will be equivalent to the
original system. This follows since the n-tuple x1 , , x n will satisfy the two equations.
32
U
LINEAR ALGEBRA- MAT 260
ai1 x1 ain x n bi
a j1 x1 a jn x n b j
ai1 x1 ain xn bi
(a j1 ai1 ) x1 (a jn ain ) xn b j bi
1.7.1 Definition
A system is said to be in strict triangular form if, in the k th equation, the coefficients of the first
k 1 variables are all zero and the coefficient of x k is non-zero (k 1, , n) . In other words, a
system of linear equations is said to be in triangular form if the number of equations is equal to
the number of unknowns and x k is the leading unknown of the k th equation. That is
a 22 x 2 a 2( n 1) x n 1 a 2 n xn b2
a n n x n bn
Example 1.9
The system
3x1 2 x 2 x3 1
33
U
LINEAR ALGEBRA- MAT 260
x 2 x3 2
2 x3 4
is in strict triangular form, since in the second equation the coefficients are 0, 1, -1, respectively,
and in the third equation the coefficients are 0, 0, 2, respectively. Because of the strict triangular
form, the system is easy to solve. It follows from the third equation that x3 2 . Using this value
in the second equation, we obtain
x2 2 2
x2 4
3x1 2 4 2 1
x1 3
34
U
LINEAR ALGEBRA- MAT 260
Unit summary
In this unit we have looked at different types of systems of linear equations. We have exposed you to
various ways of solving systems of linear equations and above all how to determine the solutions of systems
of linear equations.
35
U
LINEAR ALGEBRA- MAT 260
Unit 2
Matrix Algebra
2.0 Introduction
Welcome to unit 2. In this unit you are going to learn about different types of matrices. You will
also be taken through matrix operations which you were introduced to in MAT 110. You will
also use matrices to solve systems of linear equations using the Gauss Jordan Elimination
method and elementary row operation. Furthermore you will also be exposed to various proofs of
theorems on matrix algebra. I ‘ m sure you are ready to embark on this ‘journey’
36
U
LINEAR ALGEBRA- MAT 260
2.1 Definition
A matrix with m rows and n columns is called an m by n matrix, or m n matrix; the pair of
numbers m, n is called its size or shape.
In the above matrix, the m horizontal n- tuples a11 , a12 , a13 , , a1n , a 21 , a 22 , a 23 , , a 2 n , ,
37
U
LINEAR ALGEBRA- MAT 260
Matrices will usually be denoted by capital letters, say A, B … and the elements of the field K by
lower case letters, say a, b …
For instance the entry of matrix A in rows i and columns j is denoted by aij .
Thus, if
0 5 4
A , then
3 9 1
a 23 1 and a12 5
The notation Aij is sometimes used interchangeably with aij . We can therefore completely
denote a matrix A by aij or a ij mn
if it is important to specify the size of A.
The element aij , called the i j- entry or i j component, appears in the i th row and j th column.
If A aij is an m n matrix and B bij is a p q matrix, then A B if and only if (iff)
In other words two matrices are equal if they have the same size and if there are corresponding
entries are equal.
Example 2.1
Let A and B be two matrices with the same size, that is the same number of rows and of
columns, say,
38
U
LINEAR ALGEBRA- MAT 260
a11 b11 ; a12 b12 ; a13 b13 ; a1n b1n and so on.
The matrix a ij 0 such that aij 0 (i 1, 2, 3, , m) , ( j 1, 2, 3, , n) is called
a zero matrix.
Example 2.2
0 0 0
A 0 0 0
0 0 0
A matrix with the same number of rows as columns is called a square matrix.
A square matrix with n rows and n columns is said to be of order n, and is called a
an n- square matrix.
Example 2.3
1 0 0
B 0 1 0
0 0 1
39
U
LINEAR ALGEBRA- MAT 260
An upper triangular matrix or simply a triangular matrix is a square matrix whose entries below
the main diagonal are all zeros; that is A aij where aij 0 whenever i j .
Example 2.4
A lower triangular matrix is a square matrix whose entries above the main diagonal are all zero;
that is A aij where aij 0 whenever i j .
Example 2.5
a11 0 0 0 0 a11
a 21 a 22 0 0 0 a 21 a 22
a a 32 a33 0 0 or a a 32 a33
31 31
a n1 an2 a n3 a nn a n1 an2 a n3 a nn
A diagonal matrix is square matrix whose non- diagonal entries are all zero.
40
U
LINEAR ALGEBRA- MAT 260
Example 2.6
a1 0 0 0 a1
0 a 22 0 0 a2
0 0 a33 0 or
a3
1
0 0 0 a4 0 a4
0 0 0 a nn an
The n- square matrix with 1’s on the main diagonal and 0’s elsewhere is called the unit or
identity matrix. It is denoted by I n or simply I.
Example 2.7
1 0 0
I3 0 1 0
0 0 1
I aij where aii 1 , aij 0 , (i j ) .
If A aij is an n n matrix then an n n matrix B such that AB BA I n is called an inverse
Theorem 2.2
Proof:
41
U
LINEAR ALGEBRA- MAT 260
PQ QP I n and Q QI n Q(PR )
(QP) R
InR
R
QR
A matrix which has an inverse is called an invertible or non- singular matrix while a matrix
which has no inverse is called a singular matrix.
Theorem 2.3
(ii) If A is an invertible matrix and c is a non zero scalar, then cA is an invertible matrix and
1 1
(cA) 1 A
c
(iii)If P and Q are invertible matrices of the same size or n n matrices, then PQ is an
(v) If A is an invertible matrix, then A n is invertible for all non- negative integers n and
( A n ) 1 ( A 1 ) n
PP 1 I n P 1 P and QQ 1 I n Q 1Q
42
U
LINEAR ALGEBRA- MAT 260
Now
( PQ ) (Q 1 P 1 ) ( PQ )Q 1 ( P 1 )
P(QQ 1 ) P 1
( PI n ) P 1
PP 1
In
An n n matrix Q is said to be orthogonal matrix if the column vectors form an orthogonal set in
n .
Theorem 2.4
Proof
It follows from the definition that an n n matrix Q is orthogonal if and only if its column
vectors satisfy
T
qi q j ij
T
However, qi q j is the (i , j ) entry of the matrix Q T Q. Thus Q is orthogonal if and only if
QT Q I .
It follows from the theorem that if Q is an orthogonal matrix, then Q is invertible and Q 1 Q T
Example 2.8
43
U
LINEAR ALGEBRA- MAT 260
cos sin
Q
sin cos
cos sin
Now Q T and
sin cos
cos sin
det Q
sin cos
cos sin
Q 1
sin cos
cos sin
Q 1 Q T . Therefore the matrix Q is orthogonal.
sin cos
An n n matrix A is called
44
U
LINEAR ALGEBRA- MAT 260
Activity 2.1
1. Which of the following matrices are symmetric, and which are skew- symmetrical.
0 1 2 2 3 1
1 2 0 2 1 0
; ; ; 1 0 3 ; 3 0 1
2 3 2 0 0 1 2 3 0 1 1 2
2 1
0
1 2 6 3 3 4
5 5 ; 1 1 1
; 5 5
2 1
2 6 3 4 3
5 5 1 1 1 5 5
2 6 3
5. Show that a product of two orthogonal matrices of the same size is an orthogonal matrix.
6. Find which of the matrices are invertible and which are singular. Find the inverses of
those which are singular
2 3 2 3
1 2 1 1 1 2 0 1 1
1 2 1 0 2 1
; 0 1 2 ; 1 2 1 ; 1 0 1 ; ;
1 3 0 0 1 1 3 1 1 1 0 1 1 4 2
3
0 0 4
1 1 1 1
2 1 0 3
3 0 2 5
1 1 1 3
45
U
LINEAR ALGEBRA- MAT 260
a 0 0
7. Find the inverse of the diagonal matrix 0 b 0 , where a, b, and c are non-zero.
0 0 c
8. Let A, B and C be invertible matrices of the same size. Show that the inverse of the
product ABC is C 1 B 1 A 1
2.4.1Definition
If A aij is an m n matrix and B bij
is a p q matrix then, A B is defined if and only if m p , and n q and then
A B (aij bij ) is the m n matrix obtained by adding the corresponding elements in A and
B.
Example 2.9
Let A and B be two matrices with the same size, that is the same number of rows and of
columns, say, m n matrices.
The sum of A and B, written A+B, is the matrix obtained by adding corresponding entries
46
U
LINEAR ALGEBRA- MAT 260
If A aij is an m n matrix and K , then A aij is the m n obtained by multiplying
Theorem 2.5
Let V be a set of all m n matrices over a field K. Then for any matrices A, B, C V and any
scalars h1 , h2 K ,
i) ( A B ) C A ( B C ) Associative
ii) A 0 A
iii) A ( A) 0
iv) A B B A commutative
v) h1 ( A B) h1 A h1 B
vi) (h1 h2 ) A h1 A h2 A
viii) 1 A A
ix) 0 A 0
x) A A 2 A
47
U
LINEAR ALGEBRA- MAT 260
2.6.1 Definition:
That is
C ij ai1b j1 a i 2 b j 2 ain b jn .
n
Cij can be represented in short form as C ij aik bkj , for each i 1, , m and each j 1, , n .
k 1
Theorem 2.6
(iii) ( B C ) A BA CA
48
U
LINEAR ALGEBRA- MAT 260
1. AB 0 does not necessarily mean that A 0 or B 0 .
That is AB AC 0
2.7.1 Definition
If A aij is an m n matrix, the transpose of A is the n m matrix denoted by A t
That is A t Bij where
Theorem 2.7
Then:
ii) ( At ) t A
v) ( A r ) t ( A t ) r
49
U
LINEAR ALGEBRA- MAT 260
Further ( A t ) 1 ( A 1 ) t
Proof:
If A is invertible then AA 1 I n A 1 A
Transposing gives
( A 1 ) t A t I n
A t ( A 1 ) t I n
And so A t is invertible.
Transposing gives
A 1 A I n AA 1
Clearly from the above also implies by the previous theorem, that is ( A t ) 1 ( A 1 ) t .
Activity 2.2
1. Given that
2
1 1 2 4 0 3
A ; B ; C 1 ;
0 3 4 1 2 3 3
Find
2. Let D x be the m square scalar matrix with diagonal elements k. Show that:
50
U
LINEAR ALGEBRA- MAT 260
ii) Show that the sum, product and scalar multiple of:
a 21 x1 a 22 x 2 a 2 n x n b2
AX B , where A is the matrix of coefficients, X is the matrix of variables and B is the matrix
of constants.
Most often we will dispense (do away) with the matrix of variables X and will simply write the
argument matrix of the system as
51
U
LINEAR ALGEBRA- MAT 260
Example 2.9.1
a) 3x y 2 z 9
x 5y z 0
2 x 10 y 3z 4
b) 3w 2 x y 2 z 12
2x y 2z 1
w x 3
Answers
3 1 2 9
A/ B 1 5 1 0
2 10 3 4
3 2 1 2 12
A / B 0 2 1 2 1
1 1 0 0 3
52
U
LINEAR ALGEBRA- MAT 260
If a row of a matrix is non-zero, we call the first non-zero entry of this row a pivot or leading
entry for this row. For instance, in example (a) above, 3, 1 and -2 are leading entries for the first
row, second row and third row respectively.
ii) For any two consecutive rows Rt and Rt 1 , either Rt 1 is all zeros or the pivot of Rt 1 is
immediately to the right of the pivot of Rt . In other words in each non-zero row, the first
non-zero entry (pivot) is in a column to the left of any leading entries below it.
The variables accompanying the pivots are called leading variables. Those variables which
are not leading variables are called free parameters.
Example 2.9.2
1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
0 0 1 2 0 0 2 2 0 0 0 0 0 0 0 1
A , B , C , D
0 0 1 2 0 0 0 3 0 0 0 1 0 0 0 0
0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0
Answers
Matrices A and C are not in echelon form because they dot not satisfy the two conditions above.
For instance matrix A does not meet condition (ii) while matrix C does not satisfy condition (i).
Matrices B and D on the other hand, are in row echelon form because they both meet the two
conditions.
53
U
LINEAR ALGEBRA- MAT 260
ii) The leading entry in each non-zero row is a 1 (called a leading 1).
1 3 0 2 0 10
0 1 0 4 5 1 0
0 0 0 1 3 2 0
0 0 0 0 0 0 1
0 0 0 0 0 0 0
ii) Use elementary row operations to reduce the argument matrix to reduced row echelon
form
iii) If the resulting system is consistent, solve for the leading variables in terms of any
remaining free variables.
Example 2.9.3
x 2 x3 5
2 x1 x 2 5 x3 1
x1 x 2 3 x3 3
Solution
54
U
LINEAR ALGEBRA- MAT 260
0 1 15
2 1 51
1 1 33
1 1 33
2 1 51
0 1 15
1 1 1 3
0 1 1 5
0 1 1 5
1 1 1 3
0 1 1 5 which is a matrix in echelon form.
0 0 0 0
x1 x 2 x3 3
x 2 x 3 5
Since we have more unknowns (variables) than equations, the system has an infinite number of
solutions. That is any value of x1 , x 2 and x3 satisfy the equation represented by the third row .
Example
1 1 1 1 x 3
0 2 1 0 y 1
0
0 1 1 z 9
0 0 0 2 w 6
55
U
LINEAR ALGEBRA- MAT 260
Solution
We observe that the matrix of coefficients is already in row-echelon form. Clearly every variable
is a leading variable, and by back substitution
2w 6
w 3
Then
zw4
z 43
z 1
Also
2 y z 1
2 y 1 1
y 1
And
x y z w 3
x 1 1 3 3
x0
x 0
y 1
Therefore the unique solution is
z 1
w 3
56
U
LINEAR ALGEBRA- MAT 260
Activity 2.3
1. State which of the following matrices is in Gauss-reduced form, and for those that are not,
explain why not.
0 1 2
2 1 3 2 1 3 1 2
0 0 0
a) 0 0 1 2 b) 0 0 0 0 c) 0
0 0 0 0 0 1 1 3 0 0
0 0 0
1 2 3 4 7
1 1 8 14
1 2 1 4 0 5 7 6 8
a) 3 4 3 0 b) c) 0
2 1 7 10 2 3 1 2 5 2 4 7
0 0 2 1 0
4 1 2 6
d) 1 5 1 3
3 4 1 3
2.8.1 Definition
An Elementary row operation or simply row operation on a matrix A is any one of the
following three types of operations:
57
U
LINEAR ALGEBRA- MAT 260
row r j of A. That is ri ri r j
Example 2.9.5
i) ri ri
1 0 0 0 0 1
0 1 0 r1 r3 0 1 0
0 0 1 1 0 0
ii) ri ri
1 0 0 1 0 0
0 1 0 r3 2r3 0 1 0
0 0 1 0 0 2
iii) ri ri r j
1 0 0 1 0 0
0 1 0 r2 2r3 r2 0 1 2
0 0 1 0 0 2
Theorem 2.7
58
U
LINEAR ALGEBRA- MAT 260
B eA EA
Proof:
i) If ri ri , then
11 12 1n
i 2 in
eA ii
m2 mn
m1
1 11 12 1n
ii i 2 in
EA eA B
1
1 m2 mn
m1
12 12 1n
11 12 1n
M i ( ) A
m1 m2 mn
ii) If ri r j , then
11 1n
i j1 in
H ij A B
j j1 1n
mi mn
iii) If ri ri r j , then
11 1n
i1 j1 in in
Aij ( ) A B
mn
mi
59
U
LINEAR ALGEBRA- MAT 260
Example 2.9.6
1 0 1 1 0 1
r2 2 r1 r2
Let A 2 1 0 0 1 2 B
1 0 0 1 0 0
Next we apply the same elementary row operation on the identity matrix I 3 . We obtain
1 0 0 1 0 0
r2 2 r r2
0 1 0 2 1 0 E
0 0 1 0 0 1
1 0 01 0 1
Now note that EA 2 1 0 2 1 0
0 0 1 1 0 0
1 0 1
0 1 2 B
1 0 0
Theorem 2.8
Every elementary matrix is invertible and the inverse of each elementary is elementary.
Proof:
1
(i) ( M i ( ) 1 M i
(ii) ( H ij ) 1 H ij
60
U
LINEAR ALGEBRA- MAT 260
1
1
1 1
M i row i
1
1 1 1
1 1
M i M i ( ) 1
1 1 1
1
= M i ( ) M i .
Proving (iii)
( H ij ) 1 H ij H ij H ij ri r j
1 1 1
1 1 1
0 0 1
1 1
1 1 1
= H ij ( H ij ) 1
Proof of (iii)
61
U
LINEAR ALGEBRA- MAT 260
1 1
1 1
ri r j ri
1 1
1 1
1
1 0
=
1
1
Example 2.9.7
1 0 0 1 0 0
r3 2 r1 r3
0 1 0 0 1 0
0 0 1 2 0 1
1 0 0
1
E A31 (2) 0 1 0
2 0 1
1 0 0 1 0 0
1
E E 0 1 0 0 1 0
2 0 1 2 0 1
1 0 0
= 0 1 0 EE 1
0 0 1
62
U
LINEAR ALGEBRA- MAT 260
2.9.1 Definition
In other words, B is row equivalent to A if B can be obtained from A by a finite number of row
operations.
Theorem 2.9
Proof:
If k 1, then B 1 A
Now B is obtained from B' by applying one further elementary row operation that is B 1 B '
Cor 2.1
If R is the reduced echelon matrix of A, then there exists a finite number of elementary matrices
'1 , k such that :
A 1 2 k R
Proof:
By theorem 2.9, there exist elementary matrices '1 , ' 2 , , ' k such that
63
U
LINEAR ALGEBRA- MAT 260
R '1 2 ' k A .
' 1 ' 1
A k '1 k 1 1 R, as required
A 1 2 k R .
Cor 2.2
Proof : (i ) (ii )
A 1 (0)
0
That is
A 1 2 k
(iii ) (i) If A 1 2 E k , where each i is an elementary matrix then since elementary
matrices are invertible, then
64
U
LINEAR ALGEBRA- MAT 260
1 1
A 1 E k 1 k 1 1
( AB) 1 B 1 A 1
Thus A is invertible.
To determine whether a matrix is invertible or not we note that if the reduced echelon matrix of
A is R and
Corollary 2.2 is very important because it gives us an explicit method for finding the inverse of
an invertible matrix. If A is non-singular, then the reduced echelon matrix of A is the identity
matrix.
If 1 , 2 , , k are the elementary matrices which correspond to the elementary row operations
1 1
A 1 , 2 k I n and A 1 k k 11 I n . That is A 1 is determined by applying the same
Example 2.9.8
2 4 6
Let A 4 5 6 , calculate A 1
3 1 2
Solution
65
U
LINEAR ALGEBRA- MAT 260
2 4 6 1 0 0
4 5 6 0 1 0
3 1 2 0 0 0
And perform the elementary row operations to obtain a reduced row echelon matrix
2 4 6 1 0 0
4 5 6 0 1 0
3 1 2 0 0 0
8 7
1
1 0 0 3 3
13 11
0 1 0 2
3 3
0 0 1 11 5
1
6 3
8 7
1
3 3
13 11
1
Since A has now been reduced A 2
3 3
11 5
1
6 3
16 14 6
11
A 26 22 12
6
11 10 6
Example 2.9.9
1 3 4
Is the matrix A given by A 2 5 7 invertible?
0 1 1
Solution
1 3 4 1 0 0
2 5 7 0 1 0 r
1 2 r1 r3
0 1 1 0 0 1
66
U
LINEAR ALGEBRA- MAT 260
1 3 4 1 0 0 r
1 3 r2 r1
0 1 1 2 1 0 r
3 r2 r3
0 1 1 0 0 1
1 0 1 5 3 0
0 1 1 2 1 0
0 0 0 2 1 1
This is as far as we can go. The matrix A cannot be reduced to the identity matrix and we
conclude that A is not invertible.
If in the row reduction of A we end up with a row of zeros, then A is not invertible.
We know that if R is the reduced echelon matrix of an m n matrix A, then there exist an
invertible matrix P such that
PA R
Thus P 1 2 k
67
U
LINEAR ALGEBRA- MAT 260
P 1 ( 2 ( k 1 ( k I m )) )
And the matrix P is obtained by applying to I m the same elementary row operations as were used
to obtain R from A.
Example 2.1.0
2 1 0 1
A 1 2 1 1
2 9 4 5
2 1 0 1 1 0 0 r
1 2 r2 r1
1 2 1 1 0 1 0
2 9 4 5 0 0 1 r
3 2 r2 r3
1
r1 r1
5
0 5 2 3 1 2 0
1 2 1 1 0 1 0
r3 r1 r3
0 5 2 3 0 2 1
2 3 1 2
0 1 0
5 5 5 5
0 2 1 1 0 1 0 r
1 r2
0 0 0 0 1 4 1
1 2 1 1 0 1 0
2 3 1 2
0 1 0 r
1 2 r1 r1
5 5 5 5
0 0 0 0 1 4 1
1 1 2 1
1 0 0
5 5 5 5
2 3 1 2
0 1 0
5 5 5 5
0 0 0 0 1 4 1
68
U
LINEAR ALGEBRA- MAT 260
2 1 1 1
0 1 0
5 5 2 1 0 1 5 5
1 2 2 3
0 1 2 1 1 0 1 ; PA= R
5 5 5 5
1 4 1 2 9 4 5 0 0 0 0
Theorem 2.9 and the fact that every m n matrix is row equivalent to an m n reduced echelon
matrix have their equivalent statements in terms of elementary column operations. This leads us
to the following theorem;
Theorem 2.9.1
B A1 2 l
B 1 2 k A
From theorem 2.9.1 we see that there exists an invertible m n matrix Q such that
B = AQ
Suppose that the leading 1s appear in columns j1 j 2 , j r then by subtracting suitable multiples
of these columns from succeeding columns and then columns interchanges
1r 0r ( n r )
N where 1r is the r r identity matrix and 0 p q is the zero
0 ( mn ) r 0 ( mr )( n r )
p q matrix.
69
U
LINEAR ALGEBRA- MAT 260
Theorem 2.9.2
The matrices P and Q are calculated as explained above and an illustration is given in the
example below
Example 2.1.1
2 1 0 1
If A 1 2 1 1 from the example above if
2 9 4 5
2 1
0
5 5
1 2
P 0 then
5 5
1 4 1
1 1
1 0
5 5
2 3
PA 0 1
5 5
0 0 0 0
1 1
1 0
5 5
0 1 2 3
5 5
0 0 0 0
70
U
LINEAR ALGEBRA- MAT 260
PAQ 3 3 3 4 3 4
1 1 1
c3 c1 c3
1 0 1 0 0 0 5
5 5
2 3 0 1 0 0
0 1 0
5 5 0 1 0 1
c 4 c1 c4
0 0 0 0 5
0 0 0 1
2
1 1 c3 c2 c3
5
1 0 0 0 1 0
5 5
2 3
0 1 0 1 0 0
3
5 5 0 0 1 0 5
c 4 c 2 c4
0 0 0 0
0 0 0 1
1 1
1 0
1 0 0 0 5 5
2 3
0 1 0 0 0 1
5 5
0 0 0 0 0 0 1 0
0 0 0 1
1 1
1 0
5 5
2 3
Q 0 1
5 5
0 0 1 0
0 0 0 1
1 0 0 0
PAQ N 0 1 0 0
0 0 0 0
2.9.2.1 Definition
Two m n matrices A and B are equivalent if invertible matrices P and Q such that PAQ B
If PAQ B, then P 1 BQ 1 A
71
U
LINEAR ALGEBRA- MAT 260
What has been proved above is that every m n A is equivalent to a matrix with simple form N,
where r denotes the number of non zero rows in the reduced echelon form of A
Example 2.1.3
Find invertible matrices P and Q such that PAQ is normal form for the matrix
2 1 1
1 1 1
A
0 1 3
1 2 4
Solution
Since A is 4 3 matrix we set an argument matrix using the associated 4 4 identity matrix and
reduce it to reduced row echelon form.
2 1 1 1 0 0 0 r
1 r4
1 1 1 0 1 0 0
0 1 3 0 0 1 0
1 2 4 0 0 0 1
1 2 4 0 0 0 1 r2 r1 r2
1 1 1 0 1 0 0
0 1 3 0 0 1 0 r
4 r1 r4
2 1 1 1 0 0 0
1 2 4 0 0 0 1 r
2 r3
0 1 3 0 1 0 1 r 3 r2 r3
r4 3r2 r4 2
0 1 3 0 0 1 0
0 3 9 1 0 0 2
72
U
LINEAR ALGEBRA- MAT 260
1 0 20 0 2 1 0 0 2 1
0 1 30 0 1 0 0 0 1 0
P
0 0 00 1 1 1 0 1 1 1
0 0 0 1 0 3 2 1 0 3 2
0 0 2
1 2 1 1 1 0 2
0 0 1 0 1 1 1 0 1 3
Thus PA
0 1 1 1 0 1 3 0 0 0
1 0 3 2 1 2 4 0 0 0
We then set an argument matrix using the associated 3 3 identity matrix and then apply the
elementary column operations.
1 0 2
1 0 0 C3 2 c1 c3
0 1 3 c3 3c 2 c3
0 1 0
0 0 0
0 0 1
0 0 0
1 0 0
1 0 2 1 0 2
0 1 0
0 1 3 Q 0 1 3
0 0 0 0 0 1
0 0 1
0 0 0
Activity 2.4
0 0 1 3 5 3 2 1
4 3 2 4 2 6
0 1 1 ; ; 5 2 ; ; 2 1 1
1 1 1 2 1 2 4 3 1 6 6
4 3 1
2. Show that the given matrices are row equivalent and find a sequence of elementary row
operations that will convert A into B
73
U
LINEAR ALGEBRA- MAT 260
1 2 3 1
i) A and B ,
3 4 1 0
2 0 1 3 1 1
ii) C 1 1 0 and 3 5 1 .
1 1 1 2 2 0
Unit Summary
In this unit we exposed you to some of the special matrices, operations on matrices such as
addition, and multiplication. We have also proved some theorems and furthermore extended the
Gauss Jordan method of solving systems of linear equations to solving augmented matrices.
Finally, the use of elementary row operations was adequately discussed.
74
U
LINEAR ALGEBRA- MAT 260
Assessment
1. Determine by inspection whether a linear system with the given augmented matrix has :
a) Unique solution
0 0 12 3 2 0 1 1 1 2 3 4 0 1 2 3 4 56
0 1 31 ; 1 2 3 1 1 ; 5 6 7 8 0 ; 6 5 4 3 21
1 0 11 2 4 6 2 0 9 10 11 12 0 7 7 7 7 77
a) No solution
b) A unique solution
i) kx 2 y 3 ii) x ky 1 iii) x 2 y 3 z 2
2 x 4 y 6 kx y 1 x yz k
2x y 4z k 2
iv) x y kz 1
x ky z 1
kx y z 2
75
U
LINEAR ALGEBRA- MAT 260
b) A unique solution
a) x y 2 z 5 b) 3x 6 y 9 z 0
3x y 7 z 22 x 4y z 6
x 3 y z 10 2 x 8 y 3z 13
Activity 1.1
Activity 1.2
4 104 73
i) x ,y ,z
3 15 15
Activity 2.1
1. Symmetrical matrices
2 3 1
1 2 1 0
; ; 3 0 1
2 3 0 1 1 1 2
Skew-symmetrical
0 1 2
0 2
; 1 0 3
2 0 2 3 0
76
U
LINEAR ALGEBRA- MAT 260
2. Activity 2.2
1.
1 0
5 1 1 13 3 18
i) ii) iii) 1 3 iv) 3 9
1 1 7 4 17 0 2 4
77
U
LINEAR ALGEBRA- MAT 260
REFERENCES
78
U
LINEAR ALGEBRA- MAT 260
Unit 2
Matrix algebra
2.0 Introduction
Welcome to unit 2. In this unit you are going to learn about different types of matrices. You
will also be taken through matrix operations which you were introduced to in MAT 110.
You will also use matrices to solve systems of linear equations using the Gauss Jordan
Elimination method and elementary row operation. Furthermore you will also be exposed
to various proofs of theorems on matrix algebra. I ‘ m sure you are ready to embark on this
‘journey’
2.1 Definition
A matrix is a rectangular array of numbers called the entries or elements.
Let K be an arbitrary field. A rectangular array of the form
a11 a12 a1n
a 21 a 22 a2n
a am2 a mn
m1
where the aij are scalars (numbers) in , is called a matrix over .
A matrix with m rows and n columns is called an m by n matrix, or m n matrix; the pair of
numbers m, n is called its size or shape.
79
U
LINEAR ALGEBRA- MAT 260
In the above matrix, the m horizontal n- tuples a11 , a12 , a13 , , a1n , a 21 , a 22 , a 23 , , a 2 n ,
, a m1 , a m 2 , a m 3 , , a mn are the rows of the matrix, and the
Thus, if
0 5 4
A , then
3 9 1
a 23 1 and a12 5
The notation Aij is sometimes used interchangeably with aij . We can therefore completely
denote a matrix A by a ij or aij mn
if it is important to specify the size of A.
80
U
LINEAR ALGEBRA- MAT 260
a zero matrix.
Example 2.2
0 0 0
A 0 0 0
0 0 0
2.3.2 Square matrix
A matrix with the same number of rows as columns is called a square matrix.
A square matrix with n rows and n columns is said to be of order n, and is called a
an n- square matrix.
Example 2.3
1 0 0
B 0 1 0
0 0 1
2.3.3 Triangular matrix
2.3.3.1 Upper triangular matrix
An upper triangular matrix or simply a triangular matrix is a square matrix whose entries
below the main diagonal are all zeros; that is A aij where aij 0 whenever i j .
81
U
LINEAR ALGEBRA- MAT 260
Example 2.4
a11 a12 a13 a1n a11 a12 a13 a1n
0 a 22 a 23 a 2 n a 22 a 23 a 2 n
0 0 a33 a3n or a33 a3n
0 0 0 0 a nn a nn
Example 2.5
a11 0 0 0 0 a11
a 21 a 22 0 0 0 a 21 a 22
a a 32 a33 0 0 or a a32 a33
31 31
a n1 an2 a n3 a nn a n1 an 2 an3 a nn
2.3.4 Diagonal matrix
A diagonal matrix is square matrix whose non- diagonal entries are all zero.
Example 2.6
a1 0 0 0 a1
0 a 22 0 0 a2
0 0 a33 0 or a3
1
0 0 0 a4 0 a4
0 0 0 a nn an
2.3.5 The unit or identity matrix
The n- square matrix with 1’s on the main diagonal and 0’s elsewhere is called the unit or
identity matrix. It is denoted by I n or simply I.
82
U
LINEAR ALGEBRA- MAT 260
Example 2.7
1 0 0
I3 0 1 0
0 0 1
I aij where aii 1 , aij 0 , (i j ) .
83
U
LINEAR ALGEBRA- MAT 260
Theorem 2.3
(vi) If A is an invertible matrix, then A 1 is invertible and ( A 1 ) 1 A .
(vii) If A is an invertible matrix and c is a non zero scalar, then cA is an invertible
1
matrix and (cA) 1 A 1
c
(viii) If P and Q are invertible matrices of the same size or n n matrices, then PQ
is an invertible and ( PQ) 1 Q 1 P 1 .
Now
( PQ ) (Q 1 P 1 ) ( PQ )Q 1 ( P 1 )
P(QQ 1
) P 1
( PI n ) P 1
PP 1
In
Q 1 P 1 ( PQ) 1 or simply ( PQ ) 1 Q 1 P 1 .
2.3.8 Orthogonal matrix
An n n matrix Q is said to be orthogonal matrix if the column vectors form an orthogonal
set in n .
Theorem 2.4
An n n matrix Q is orthogonal if and if Q T Q I
Proof
It follows from the definition that an n n matrix Q is orthogonal if and only if its column
vectors satisfy
T
qi q j ij
84
U
LINEAR ALGEBRA- MAT 260
T
However, qi q j is the (i , j ) entry of the matrix Q T Q. Thus Q is orthogonal if and only if
QT Q I .
It follows from the theorem that if Q is an orthogonal matrix, then Q is invertible and
Q 1 Q T
Example 2.8
Consider the matrix Q below
cos sin
Q
sin cos
cos sin
Now Q T and
sin cos
cos sin
det Q
sin cos
cos sin
Q 1
sin cos
You can see that
cos sin
Q 1 Q T . Therefore the matrix Q is orthogonal.
sin cos
2.3.9 Idempotent matrix
An n n matrix A is said to be idempotent if and only if A 2 A .
85
U
LINEAR ALGEBRA- MAT 260
Activity 2.1
2. Which of the following matrices are symmetric, and which are skew- symmetrical.
0 1 2 2 3 1
1 2 0 2 1 0
; ; ; 1 0 3 ; 3 0 1
2 3 2 0 0 1 2 3 0 1 1 2
2. Let A be any square matrix. Show that A AT is a symmetric matrix
3. Show that the products AT A and AAT are symmetric matrices.
4. Show that the following matrices are orthogonal.
2 1
0
1 2 6 3 3 4
5 5 ; 1 1 1
; 5 5
2 1
2 6 3 4 3
5 5 1 1 1 5 5
2 6 3
10. Show that a product of two orthogonal matrices of the same size is an orthogonal
matrix.
11. Find which of the matrices are invertible and which are singular. Find the inverses
of those which are singular
2 3 2 3
1 2 1 1 1 2 0 1 1
1 2 1 0 2 1
; 0 1 2 ; 1 2 1 ; 1 0 1 ; ;
1 3 0 0 1 1 3 1 1 1 0 1 1 4 2
3 0 0 4
1 1 1 1
2 1 0 3
3 0 2 5
1 1 1 3
a 0 0
12. Find the inverse of the diagonal matrix 0 b 0 , where a, b, and c are non-zero.
0 0 c
13. Let A, B and C be invertible matrices of the same size. Show that the inverse of the
product ABC is C 1 B 1 A 1
86
U
LINEAR ALGEBRA- MAT 260
The sum of A and B, written A+B, is the matrix obtained by adding corresponding entries
a11 a12 a1n b11 b22 1n
a a 22 a 2 n b21 b22 2n
A B 21
a am2 a mn bm1 bm 2 bmn
m1
a11 b11 a12 b12 a1n b1n
a b21 a 22 b22 a 2 n b2 n
A B 21
a b a m 2 b1m a mn bmn
m1 m1
87
U
LINEAR ALGEBRA- MAT 260
xii) A 0 A
xiii) A ( A) 0
xiv) A B B A commutative
xv) h1 ( A B) h1 A h1 B
xvi) (h1 h2 ) A h1 A h2 A
xvii) (h1 h2 ) A h1 (h2 A)
xviii) 1 A A
xix) 0A 0
xx) A A 2 A
2.6 Matrix Multiplication
2.6.1 Definition:
If A aik is an m n matrix and B bkj is an n r matrix, then the product AB is the
m r matrix C cij .
The i, j entry of the product is computed as follows; multiplying elements in the i th row
of A with the corresponding elements in the j th column of B and summing up. That is
C ij ai1b j1 a i 2 b j 2 ain b jn .
n
Cij can be represented in short form as C ij aik bkj , for each i 1, , m and
k 1
each j 1, , n .
88
U
LINEAR ALGEBRA- MAT 260
Theorem 2.7
Properties of the transpose
If A and B are m n and p q matrices respectively, and k be a scalar.
Then:
vi) ( A B ) t At B t where both sides are defined if and only if m p and
nq
vii) ( At ) t A
x) ( A r ) t ( A t ) r
89
U
LINEAR ALGEBRA- MAT 260
Transposing gives
( A 1 ) t A t I n
A t ( A 1 ) t I n
And so A t is invertible.
Conversely, if A t is invertible, then A t ( A t ) 1 I n ( A t ) 1 A t
Transposing gives
A 1 A I n AA 1
Clearly from the above also implies by the previous theorem, that is ( A t ) 1 ( A 1 ) t .
Activity 2.2
3. Given that
2
1 1 2 4 0 3
A ; B ; C 1 ;
0 3 4 1 2 3 3
Find
ii) A + B ii) 3A – 4B iii) AT iv) C T AT v) A 3
4. Let D x be the m square scalar matrix with diagonal elements k. Show that:
a 21 x1 a 22 x 2 a 2 n x n b2
a m1 x1 a m 2 x 2 a mn x n bm can be written in matrix form as:
90
U
LINEAR ALGEBRA- MAT 260
Example 2.9.1
Write the argument matrix of the systems given
a) 3x y 2 z 9
x 5y z 0
2 x 10 y 3z 4
b) 3w 2 x y 2 z 12
2x y 2z 1
w x 3
Answers
c) The argument matrix is
3 1 2 9
A/ B 1 5 1 0
2 10 3 4
91
U
LINEAR ALGEBRA- MAT 260
The variables accompanying the pivots are called leading variables. Those variables
which are not leading variables are called free parameters.
Example 2.9.2
State whether the given matrices are in echelon form or not.
1 0 1 1 1 0 1 1 1 0 1 1 1 0 1 1
0 0 1 2 0 0 2 2 0 0 0 0 0 0 0 1
A , B , C , D
0 0 1 2 0 0 0 3 0 0 0 1 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Answers
Matrices A and C are not in echelon form because they dot not satisfy the two conditions
above. For instance matrix A does not meet condition (ii) while matrix C does not satisfy
condition (i).
Matrices B and D on the other hand, are in row echelon form because they both meet the
two conditions.
92
U
LINEAR ALGEBRA- MAT 260
2 x1 x 2 5 x3 1
x1 x 2 3 x3 3
Solution
We begin by writing the argument matrix as follows:
93
U
LINEAR ALGEBRA- MAT 260
0 1 15
2 1 51
1 1 33
1 1 33
2 1 51
0 1 15
1 1 1 3
0 1 1 5
0 1 1 5
1 1 1 3
0 1 1 5 which is a matrix in echelon form.
0 0 0 0
x 2 x 3 5
Since we have more unknowns (variables) than equations, the system has an infinite
number of solutions. That is any value of x1 , x2 and x3 satisfy the equation represented by
the third row .
Example
Solve the system of linear equations
1 1 1 1 x 3
0 2 1 0 y 1
0
0 1 1 z 9
0 0 0 2 w 6
Solution
We observe that the matrix of coefficients is already in row-echelon form. Clearly every
variable is a leading variable, and by back substitution
2w 6
94
U
LINEAR ALGEBRA- MAT 260
w 3
Then
zw4
z 43
z 1
Also
2 y z 1
2 y 1 1
y 1
And
x y z w 3
x 1 1 3 3
x0
x 0
y 1
Therefore the unique solution is
z 1
w 3
Given a matrix A M mn (F ) , by following Gauss Jordan reduction, we can find a matrix
P Glm ( f ) such that PA B is in row-echelon form
Activity 2.3
1. State which of the following matrices is in Gauss-reduced form, and for those that are
not, explain why not.
0 1 2
2 1 3 2 1 3 1 2
0 0 0
a) 0 0 1 2 b) 0 0 0 0 c) 0
0 0 0 0 0 1 1 3 0 0
0 0 0
2. Find a Gauss- reduced form for each of the following matrices.
95
U
LINEAR ALGEBRA- MAT 260
1 2 3 4 7
1 1 8 14
1 2 1 4 0 5 7 6 8
a) 3 4 3 0 b) c) 0
2 1 7 10 2 3 1 2 5 2 4 7
0 0 2 1 0
4 1 2 6
d) 1 5 1 3
3 4 1 3
An Elementary row operation or simply row operation on a matrix A is any one of the
following three types of operations:
(iv) Interchange of two rows of A. That is ri r j
1 0 0 0 0 1
0 1 0 r1 r3 0 1 0
0 0 1 1 0 0
v) ri ri
1 0 0 1 0 0
0 1 0 r3 2r3 0 1 0
0 0 1 0 0 2
vi) ri ri r j
96
U
LINEAR ALGEBRA- MAT 260
1 0 0 1 0 0
0 1 0 r2 2r3 r2 0 1 2
0 0 1 0 0 2
Theorem 2.7
If an m n matrix B can be obtained from an m n matrix A by applying an elementary
row operation, then B is equal to the product of the corresponding m n elementary matrix
A. If e is the elementary row operation, then
B eA EA
Proof:
We shall prove the theorem by considering each separate type.
iv) If ri ri , then
11 12 1n
i 2 in
eA ii
m2 mn
m1
1 11 12 1n
ii i 2 in
EA eA B
1
1 m1 m2 mn
12 12 1n
11 12 1n
M i ( ) A
m1 m2 mn
v) If ri r j , then
11 1n
i j1 in
H ij A B
j j1 1n
mi mn
97
U
LINEAR ALGEBRA- MAT 260
vi) If ri ri r j , then
11 1n
i1 j1 in in
Aij ( ) A B
mn
mi
Example
1 0 1 1 0 1
r2 2 r1 r2
Let A 2 1 0 0 1 2 B
1 0 0 1 0 0
Next we apply the same elementary row operation on the identity matrix I 3 . We obtain
1 0 0 1 0 0
r2 2 r r2
0 1 0 2 1 0 E
0 0 1 0 0 1
1 0 01 0 1
Now note that EA 2 1 0 2 1 0
0 0 1 1 0 0
1 0 1
0 1 2 B
1 0 0
Theorem 2.8
Every elementary matrix is invertible and the inverse of each elementary is elementary.
Proof:
We shall consider each separate type by verifying that
1
(iv) ( M i ( ) 1 M i
(v) ( H ij ) 1 H ij
98
U
LINEAR ALGEBRA- MAT 260
1
= M i ( ) M i .
Proving (iii)
( H ij ) 1 H ij H ij H ij ri r j
1 1 1
1 1 1
0 0 1
1 1
1 1 1
= H ij ( H ij ) 1
Proof of (iii)
( Aij ( )) 1 Aij ( ) Aij ( ) Aij ( )
99
U
LINEAR ALGEBRA- MAT 260
1 1
1 1
ri r j ri
1 1
1 1
1
1 0
=
1
1
= Aij ( ) Aij ( )
Example
1 0 0 1 0 0
r3 2 r1 r3
0 1 0 0 1 0
0 0 1 2 0 1
1 0 0
1
E A31 (2) 0 1 0
2 0 1
1 0 0 1 0 0
1
E E 0 1 0 0 1 0
2 0 1 2 0 1
1 0 0
= 0 1 0 EE 1
0 0 1
2.9 Row equivalent matrices
2.9.1 Definition
A matrix B is row equivalent to matrix A if there exists a finite sequence E1 , E 2 , , E k of
elementary matrices such that B E k E k 1 E1 A .
100
U
LINEAR ALGEBRA- MAT 260
Proof:
B is row equivalent to A B can be obtained from A by finite, say k sequences of
elementary row operations. We shall prove the theorem by induction.
If k 1, then B 1 A
A 1 2 k R
Proof:
By theorem 2.9, there exist elementary matrices '1 , ' 2 , , ' k such that
R '1 2 ' k A .
' ' '
But 1 , 2 , , k is non-singular and its inverse is also an elementary matrix. Thus
' 1 ' 1
A k '1 k 1 1 R, as required
A 1 2 k R .
Cor 2.2
The following matrices are equivalent.
iv) The n n matrix A is invertible
v) The homogeneous system AX 0 of n linear equations in n variable has no trivial
solution
vi) A is a product of elementary matrices.\
101
U
LINEAR ALGEBRA- MAT 260
Proof : (i ) (ii )
If A is invertible, the system of n equations in n variables given by AX 0 has non-trivial
solution since X ( A 1 A) X A 1 ( AX )
A 1 (0)
0
( AB) 1 B 1 A 1
Thus A is invertible.
To determine whether a matrix is invertible or not we note that if the reduced echelon
matrix of A is R and
iii) If R I n , the matrix A is invertible
iv) If R contains a row of zeros, then A is not invertible.
Corollary 2.2 is very important because it gives us an explicit method for finding the
inverse of an invertible matrix. If A is non-singular, then the reduced echelon matrix of A
is the identity matrix.
If 1 , 2 , , k are the elementary matrices which correspond to the elementary row
operations which must be performed on A to give I n . That is I n k k 1 1 A , then
1 1
A 1 , 2 k I n and A 1 k k 11 I n . That is A 1 is determined by applying the
same elementary row operations to I n as were used to obtain I n from A.
Example
102
U
LINEAR ALGEBRA- MAT 260
2 4 6
Let A 4 5 6 , calculate A 1
3 1 2
Solution
We first put I next to A in an augmented matrix form as follows
2 4 6 1 0 0
4 5 6 0 1 0
3 1 2 0 0 0
And perform the elementary row operations to obtain a reduced row echelon matrix
2 4 6 1 0 0
4 5 6 0 1 0
3 1 2 0 0 0
8 7
1
1 0 0 3 3
13 11
0 1 0 2
3 3
0 0 1 11 5
1
6 3
8 7
1
3 3
13 11
1
Since A has now been reduced A 2
3 3
11 5
1
6 3
16 14 6
1
1
A 26 22 12
6
11 10 6
Example
1 3 4
Is the matrix A given by A 2 5 7 invertible?
0 1 1
Solution
103
U
LINEAR ALGEBRA- MAT 260
1 3 4 1 0 0
2 5 7 0 1 0 r
1 2 r1 r3
0 1 1 0 0 1
1 3 4 1 0 0 r1 3 r2 r1
r3 r2 r3
0 1 1 2 1 0
0 1 1 0 0 1
1 0 1 5 3 0
0 1 1 2 1 0
0 0 0 2 1 1
This is as far as we can go. The matrix A cannot be reduced to the identity matrix and we
conclude that A is not invertible.
If in the row reduction of A we end up with a row of zeros, then A is not invertible.
P 1 ( 2 ( k 1 ( k I m )) )
And the matrix P is obtained by applying to I m the same elementary row operations as
were used to obtain R from A.
Example
104
U
LINEAR ALGEBRA- MAT 260
2 1 0 1
A 1 2 1 1
2 9 4 5
2 1 0 1 1 0 0 r
1 2 r2 r1
1 2 1 1 0 1 0
2 9 4 5 0 0 1 r
3 2 r2 r3
1
r1 r1
5
0 5 2 3 1 2 0
1 2 1 1 0 1 0
r3 r1 r3
0 5 2 3 0 2 1
2 3 1 2
0 1 0
5 5 5 5
0 2 1 1 0 1 0 r
1 r2
0 0 0 0 1 4 1
1 2 1 1 0 1 0
2 3 1 2
0 1 0 r
1 2 r1 r1
5 5 5 5
0 0 0 0 1 4 1
1 1 2 1
1 0 0
5 5 5 5
2 3 1 2
0 1 0
5 5 5 5
0 0 0 0 1 4 1
2 1 1 1
0 1 0
5 5 2 1 0 1 5 5
1 2 2 3
0 1 2 1 1 0 1 ; PA= R
5 5 5 5
1 4 1 2 9 4 5 0 0 0 0
105
U
LINEAR ALGEBRA- MAT 260
Theorem 2.9 and the fact that every m n matrix is row equivalent to an m n reduced
echelon matrix have their equivalent statements in terms of elementary column operations.
This leads us to the following theorem;
Theorem 2.9.1
If an m n matrix A is column equivalent to an m n matrix B, then there exist a finite
number of elementary matrices 1 , 2 , , l such that
B A1 2 l
B 1 2 k A
From theorem 2.9.1 we see that there exists an invertible m n matrix Q such that
B = AQ
Let us now apply this to the reduced echelon matrix R of an m n matrix A.
Suppose that the leading 1s appear in columns j1 j 2 , j r then by subtracting suitable
multiples of these columns from succeeding columns and then columns interchanges
ci c j , c 2 cj 2 ,, cr cjr , we have that R is column equivalent to the matrix
1r 0r ( n r )
N where 1r is the r r identity matrix and 0 p q is the zero
0 ( mn ) r 0 ( mr )( n r )
p q matrix.
Theorem 2.9.2
If A is an m n then there exist an invertible m m matrix P and an invertible m n matrix
Q such that PAQ N where N is the matrix given above.
The matrices P and Q are calculated as explained above and an illustration is given in the
example below
Example
2 1 0 1
If A 1 2 1 1 from the example above if
2 9 4 5
2 1
0
5 5
1 2
P 0 then
5 5
1 4 1
106
U
LINEAR ALGEBRA- MAT 260
1 1
1 0
5 5
2 3
PA 0 1
5 5
0 0 0 0
Now we apply the elementary column operations on the argument matrix.
1 1
1 0
5 5
0 1 2 3
5 5
0 0 0 0
PAQ 3 3 3 4 3 4
1 1 1
c3 c1 c3
1 0 1 0 0 0
5
5 5
2 3 0 1 0 0
0 1 0
5 5 0 1 0 1
c 4 c1 c4
0 0 0 0 5
0 0 0 1
2
1 1 5
c3 c2 c3
1 0 0 0 1 0
5 5
2 3
0 1 0 1 0 0
3
5 5 0 0 1 0 5
c 4 c 2 c4
0 0 0 0
0 0 0 1
1 1
1 0
1 0 0 0 5 5
2 3
0 1 0 0 0 1
5 5
0 0 0 0 0 0 1 0
0 0 0 1
1 1
1 0
5 5
2 3
Q 0 1
5 5
0 0 1 0
0 0 0 1
107
U
LINEAR ALGEBRA- MAT 260
1 0 0 0
PAQ N 0 1 0 0
0 0 0 0
2.9.2 Equivalent Matrices
2.9.2.1 Definition
Two m n matrices A and B are equivalent if invertible matrices P and Q such that
PAQ B
If PAQ B, then P 1 BQ 1 A
PAQ B and P BQ C then ( P P ) A(QQ ) C
What has been proved above is that every m n A is equivalent to a matrix with simple
form N, where r denotes the number of non zero rows in the reduced echelon form of A
N may be regarded as a canonical form of A under this equivalence relation. That is a
representative of the equivalence class which contains A under this equivalence relation.
Later it will be shown that r is the rank of the matrix A and N is called the normal form of
the matrix A.
Example
Find invertible matrices P and Q such that PAQ is normal form for the matrix
2 1 1
1 1 1
A
0 1 3
1 2 4
Solution
Since A is 4 3 matrix we set an argument matrix using the associated 4 4 identity matrix
and reduce it to reduced row echelon form.
2 1 1 1 0 0 0 r
1 r4
1 1 1 0 1 0 0
0 1 3 0 0 1 0
1 2 4 0 0 0 1
1 2 4 0 0 0 1 r
2 r1 r2
1 1 1 0 1 0 0
0 1 3 0 0 1 0 r
4 r1 r4
2 1 1 1 0 0 0
108
U
LINEAR ALGEBRA- MAT 260
1 2 4 0 0 0 1 r 2 r3
r3 r2 r3
0 1 3 0 1 0 1
r4 3r2 r4 2
0 1 3 0 0 1 0
0 3 9 1 0 0 2
1 0 2 0 0 2 1 0 0 2 1
0 1 3 0 0 1 0 0 0 1 0
P
0 0 0 0 1 1 1 0 1 1 1
0 0 0 1 0 3 2 1 0 3 2
0 0 2 1 2 1 1 1 0 2
0 0 1 0 1 1 1 0 1 3
Thus PA
0 1 1 1 0 1 3 0 0 0
1 0 3 2 1 2 4 0 0 0
We then set an argument matrix using the associated 3 3 identity matrix and then apply
the elementary column operations.
1 0 2
1 0 0 C3 2 c1 c3
0 1 3 c3 3c 2 c3
0 1 0
0 0 0
0 0 1
0 0 0
1 0 0
1 0 2 1 0 2
0 1 0
0 1 3 Q 0 1 3
0 0 0 0 0 1
0 0 1
0 0 0
Activity 2.4
3. Use elementary row operations to reduce the given matrix to:
c) Row echelon form
d) Reduced echelon form
0 0 1 3 5 3 2 1
4 3 2 4 2 6
0 1 1 ; ; 5 2 ; ; 2 1 1
1 1 1 2 1 2 4 3 1 6 6
4 3 1
4. Show that the given matrices are row equivalent and find a sequence of elementary
row operations that will convert A into B
109
U
LINEAR ALGEBRA- MAT 260
1 2 3 1
iii) A and B ,
3 4 1 0
2 0 1 3 1 1
iv) C 1 1 0 and 3 5 1 .
1 1 1 2 2 0
Unit Summary
In this unit we exposed you to some of the special matrices, operations on matrices such as
addition, and multiplication. We have also proved some theorems and furthermore
extended the Gauss Jordan method of solving systems of linear equations to solving
augmented matrices. Finally, the use of elementary row operations was adequately
discussed.
110
U
LINEAR ALGEBRA- MAT 260
Assessment
6. Determine by inspection whether a linear system with the given augmented matrix
has :
d) Unique solution
e) Infinitely many solutions
f) No solution. Justify your answers
0 0 12 3 2 0 1 1 1 2 3 4 0 1 2 3 4 56
0 1 31 ; 1 2 3 1 1 ; 5 6 7 8 0 ; 6 5 4 3 21
1 0 11 2 4 6 2 0 9 10 11 12 0 7 7 7 7 77
2x y 4z k 2
iv) x y kz 1
x ky z 1
kx y z 2
8. Give examples of homogeneous systems of m linear equations in n unknowns or
variables with m = n and with m > n that have
c) Infinitely many solutions
d) A unique solution
9. Use gauss elimination for augmented matrices to solve
b) x y 2 z 5 b) 3x 6 y 9 z 0
3x y 7 z 22 x 4y z 6
x 3 y z 10 2 x 8 y 3z 13
10. Prove that if A is invertible and AB = 0, then B = 0.
111
U
LINEAR ALGEBRA- MAT 260
112
U
LINEAR ALGEBRA- MAT 260
REFERENCES
Poole D; (200) Linear Algebra: A Modern Introduction
Pacific Groove: Thomson Learning.
Finkbeiner II D.T; (1972) Elements of Linear Algebra
San Franscisco: W.H Freeman & Co.
Hamilton A.G; (1989) Linear Algebra: An introduction with concurrent examples.
New York: Cambridge University Press
James W.D &Noble B; (1988) Applied Linear Algebra
New Jersey: Prentice-Hall
Leon S.J ; (2010) Linear Algebra with applications
New jersey: Pearson Education inc.
113