Chapter 1: Matrix Algebra and System of Linear Equations

1.1 Definition of a matrix

Definition: A rectangular array of numbers arranged in horizontal rows and vertical columns is
called a matrix.

( , [ ] ( )

The numbers in the row and column are called the elements or entries of the matrix.
The size or order of a matrix is indicated by the number of rows and columns. A matrix with m
rows and n columns is said to be of dimension (or order) m by n and denoted by m  n matrix.
Matrices are denoted by upper case (capital) letters and the standard notation for the elements is
to use the same letter but with lower case and the two subscripts which describe the position of
the elements. In the above matrix the first subscript refers to the position of rows and second
subscript refers to the position of columns to which the element belongs. Thus, the above matrix
is denoted by ( ) for i  1, 2, 3,  , m and j  1, 2, 3,  , n . In the subscript the first

letter m denotes the number of rows and the second letter n denotes the number of columns. The

element in the row i and column j is called the (i, j ) th entry and often written as aij .

Example: The following arrangements of numbers give the distance between the indicted cities.

Addis Ababa Dire Dawa Arba Minch Mekele

AddisAbaba  0 505 515 710 
 
 
 505 0 600 1215 
Dire Dawa  
 
 515 600 0 1225 
Arba Minch 
 
 
Mekele  710 1215 1225 0 

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 1

Basic definitions in matrix

1. A matrix with one row and n  columns is called a row matrix.

For example the matrix is a row matrix.

2. A matrix with one columns and m  rows is called a column matrix.

For example the matrix ( ) is a column matrix.

3. A matrix with all its elements zero is call a zero (or null) matrix.

For example the matrix ( + is a zero matrix.

The matrix ( + is also a zero matrix.

4. A matrix with the number of rows m equal to the number of columns n is called a square
matrix of order n . For square matrix, the diagonal from the top left to the bottom right corner is
called the principal (or main) diagonal of the matrix.

For example ( , is a square matrix of order n .

The matrix ( ) is square matrix of order two.

The matrix ( + is square matrix of order three.

If the matrix is not square it is called a rectangular matrix of dimension m  n .

5. A square matrix ( ) is called a diagonal matrix if all the elements off the main

diagonal are zero, that is, for all .

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 2

  For example ( , is a diagonal matrix of order n .

 The matrix ( ) is a diagonal matrix of order two.

 The matrix ( + is diagonal matrix of order three.

6. A diagonal matrix of order n who’s all the diagonal elements are one is called a unit (or
identity) matrix and denoted by .

 For example the matrix ( ) is a unit matrix of order two.

 The matrix ( + is unit matrix of order three.

 The matrix ( ) is unit matrix of order n .

7. A square matrix is called an upper-triangular matrix if all the elements below the principal
diagonal are zero. On the other hand, if all the elements above the principal diagonal are zero,
then it is called a lower-triangular matrix.

 For example the matrix ( + is an upper triangular matrix.

 The matrix ( + is a lower triangular matrix.

In general; an upper triangular matrix or lower triangular matrices are known as triangular
matrix.

9. The trace of a square matrix is the sum of its diagonal elements.

For example for a matrix ( + , trace of , denoted by

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 3

Definition: Two matrices ( ) and ( ) of the same dimensions are said to

be equal matrices if and only if for every and for every

.

1.2 Operation with matrices

1.2.1 Addition of matrices

Addition of two matrices would be possible if and only if the two matrices have the same
dimension. In this case if ( ) and ( ) then ( ) .

Example: Find the sum of the matrices ( + and ( +

Solution: Adding the components with the terms in the two matrices, we have

( + ( + ( + ( +.

Properties of matrix addition

Let be m  n matrices, then

a) (Addition is commutative)

b) (Addition is associative)

c) There is a unique m  n zero matrix such that for any matrix .

The matrix is called the additive identity element.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 4

1.2.2 Scalar multiplication of a matrix

Scalar multiplication is the product of a real number by a matrix, which is completely different
from product of two matrices.

The product of a number with the matrix of dimension , the matrix denoted by , is a
matrix obtained by multiplying each entry of A by . The dimension of the resulting matrix is

also . That is

( ,

Example: Let ( + , and ( + , then find ⁄ .

Solution: multiplying the components with the terms in the matrix, we have

⁄ ⁄
( + ( ⁄ +
⁄

Note: The Subtraction of two matrices of the same dimension ( ) ( )

can be defined by:

( )

( + ( + ( + ( + ( +

( +.

Properties of scalar multiplication

Let and be matrices with real numbers and then

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 5

 a)

b)

c) For any matrix of dimension there is a unique matrix such

that .The matrix is the zero matrix of dimension .

1.2.3 Multiplication of matrices

The multiplication of two matrices is a little bit more complicated than the sum or difference of two
matrices. Any pair of matrices cannot be multiplied together and the order in which matrices are
multiplied can matter.

Let ( , ( ,

We can find the product of two matrices A and B , denoted by AB , if and only if

i) the number of columns of and the number of rows of are equal (that is, is
defined, if dimension of is and dimension of is ).
ii) The dimension of the resulting matrix is determined by the number of rows of
and the number of columns of (that is, the dimension of is ).
iii) To obtain the entry of matrix , we take the dot product of the row of
matrix A with the column of matrix . That is

Example: Let ( ) and ( + , then find and whenever possible.

Solution: To find the product write the two matrices adjacent to each other and perform the
operations. The product is defined as the number of column of equal to the number of rows
of , that is 3. Also the resulting matrix is of order 2.
Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 6
 ( )( + ( * ( )

However, the product is also defined as the number of column of is 2 while the number of
rows of is also2, thus the resulting matrix is also of order 3 as shown below.

( +( ) ( )

( +

From this example we can observe that, matrix multiplication is not commutative.

Example: Let ( + and ( +, then find and if it is possible.

Solution: Matrix is of dimension and matrix is of dimension , the number of

columns of is equal to the number of rows of matrix , thus the product is defined and its
dimension is . Thus we can find:

( +( + ( )

( +

On the other hand, the number of columns of and the number of rows of is not equal, thus
the product is not possible.

Properties of multiplication of matrices

a) Let be matrix, be matrix and be matrix, then

(Matrix multiplication is associative)

b) Let and be matrix, and be matrix, then

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 7

 (Multiplication is right distributive over addition)

c) Let and be matrix, and be matrix, then

(Multiplication is left distributive over addition)

d) Let be matrix, then

1.2.4 Transpose of a matrix

The transpose of matrix is the matrix obtained by interchanging the rows and
columns of the given matrix and denoted by or .

Let ( ,, then ( ,.

For example given matrix ( +, it’s transpose will be ( +

 A square matrix is said to be symmetric if its transpose is equal to the matrix, that is .
 A square matrix is said to be skew symmetric if its transpose is equal to the negative of ,
that is .

For example, If ( +, then ( +. Thus and is symmetric matrix.

If ( +, then ( +. Thus and is skew symmetric matrix.

Note Every diagonal matrix is symmetric matrix.

Properties of transpose of a matrix

Let A be matrix and B be matrix with real numbers r then

a)
Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 8
 b)

c)

1.2.5 Inverse of a matrix

Let be a square matrix of order . We say that is invertible or non-singular if there exists a
square matrix of order such that

In this case is called the inverse of . Otherwise, is called singular or noninvertible.

Remark a) If the inverse of a matrix exist, then it is unique.

b) The unique matrix where it exist is called the inverse of , and denoted by .

c) does not mean .

Example: Find the inverse of the square matrix ( ), using the definition of inverse of a

matrix.

Solution: Consider a square matrix of order two, let ( ) such that , then

( )( ) ( )

( * ( )

From equation we have and put this in equation

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 9

 ⁄ ⁄

From equation we have and put this in equation

⁄ ⁄

⁄ ⁄
Therefore the inverse of matrix is the matrix ( *
⁄ ⁄

To check that it also holds

⁄ ⁄
( *( ) ( , ( )
⁄ ⁄

It also holds as we have. This is a system of four linear equations with four unknowns and we can
also split the system in to two simultaneous equations

To finding the inverse of a matrix we can use two techniques determinant of a matrix and
elementary raw operations.

1.2.6 Elementary raw operations

Elementary raw operations involves the following operations on a given raw of a matrix.

i) Interchange any two rows of a matrix.

ii) Multiply each element in a row of a matrix by the same nonzero number.
iii) Change a row of a matrix by adding to it a constant multiple of another row.
Definition A row of a matrix is said to have k-leading zeros if the first -elements of the row all
zero and the element of the row is not zero.

For example consider the matrix ( +

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 10

  The first row has no leading zero (has zero leading zero).
 The second row has one leading zero.
 The third has two leading zero.

Definition A matrix is said to be in the row-echelon form, if

i) the first non-zero entry in each row is 1.

ii) row has k-leading zeros , then the number of leading zeros entries in row
is greater than the number of leading zero entries in row .
iii) there are rows whose entries are all zero, they are below the rows having non-
zero entries
For example the following matrices are in the row-echelon form:

1 5 7  6 
 
( +, ( +,  0 1 7  4
0 0 0 1 
 

Definition A matrix is said to be in reduced row-echelon form, if

i) it is in row-echelon form and

ii) a column containing the first nonzero entry of some row, then all other
entries in that column are zero.

Example: ( +, ( +

Example: Reduce the matrix ( ), in its row-echelon form.

Solution: To reduce the given matrix into its row-echelon form we will use the elementary row
operation. Observe on the first hand each row has zero leading zero, thus to reduce the given
matrix into row echelon form the position of the element in the first row and first column have to
be one. Following this observations all the elements below the element in the first row and first
column have to be zero as we wish to increase the number of leading zeros as we move down the
rows.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 11

To change a nonzero element to one, multiply the entire row with its reciprocal. To change the

element 2 to one multiply the first row by . That is the equation is applied on the first

row.

To change a nonzero element of a row to zero subtract from this row the ratio of the element that
becomes zero in the row by the corresponding element that becomes one in its column and
multiply with the corresponding row that becomes one. To change the element 5 to zero from row

two subtract the ratio multiplied by the first row . That is the equation is applied

on the second row.

( ) ( ) ( )

Definition An matrix is said to be row equivalent to matrix B, if results from by

a finite sequence of elementary row operations.

In the above example ( ) is row-equivalent to ( ) and ( ) is also row

equivalent to ( ). Furthermore, ( ) is also row equivalent to ( )

Definition: The rank of a matrix is the number of non-zero rows in its row-echelon form and
denoted by

For example, the rank of the above matrix in example is 2.

3 5 1 4 
 
Example: Reduce the matrix B   2  1 1 1  , in its row-echelon form and find its rank.
8 9 3 9
 

Solution: To find the rank of a given matrix, we reduce the given matrix in the row-echelon form
by using elementary row operations and decide its rank.

3 5 1 4  1 5 3 13 4 3 1 5 3 1 3 4 3
     
2 1 1 1   0  13 3 1 3  5 3  0  13 3 1 3  5 3 
8 9 3 9  0  13 3 1 3  5 3  0 0 
    0 0

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 12

The rank of matrix is 2.

1.2.7 Inverse of a matrix using augmentation method

In finding the inverse of a square matrix of order using the elementary row operation;
augment the given matrix with the corresponding identity matrix; that is construct a rectangular
matrix whose dimension will be by attaching on the right side of an identity
matrix of order . Then use elementary row operation and reduce the matrix to the form
of which is always possible if is invertible matrix. Finally we take

Let ( ), then ( | ),

by using elementary row operation one can transform this matrix to the form of

( | ,.

Therefore, ( ,

Example: Find the inverse of the matrix ( ), using the augmented method.

Solution: Using the identity matrix of order 2 the augmented matrix will be:

( | )→ ( | )→ ( | *
⁄ ⁄ ⁄ ⁄

Therefore, ( *
⁄ ⁄

Example: Find the inverse of the matrix ( +, using the augmented method.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 13

Solution: Using the identity matrix of order 2 the augmented matrix will be:

⁄ ⁄ ⁄ ⁄ ⁄
( | + ⁄ ( ⁄ ⁄ | ⁄ + ⁄
⁄ ⁄ ⁄ ⁄ ⁄

⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄
( ⁄ | ⁄ ⁄ + ⁄ ( | ⁄ ⁄ ⁄ +
⁄ ⁄ ⁄ ⁄ ⁄ ⁄ ⁄

⁄ ⁄ ⁄
Therefore, ( ⁄ ⁄ ⁄ +
⁄ ⁄ ⁄

1.2.8 System of linear equations

The general system of linear of -equations with -unknowns is given by

where and are given real numbers for and . is the

coefficient of the unknown in the equation . A solution set of the system (2.1) is an -tuple of
real numbers which satisfies each of the -equations in (2.1).

The system (2.1) above can be written in matrix notation with the components as:

( ,, called the coefficient matrix, ( ,, called the variable

matrix and ( ,, called the constant term matrix. Therefore, the system (2.1) can be

written as

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 14

For example; to express the system of linear equation { using the matrix

notation we have: ( +( + ( +

The general system of linear of -equations with -unknowns is given by (1.1) or (1.2) is said to be
inconsistent if it has no solution, otherwise it is said to be consistent; that is, it has one or many
solutions.

If , then the system (1.1) or (1.2) is said to be homogenous system and if , is system
(1.1) or (1.2) is said called non-homogenous system.

A homogenous system of linear equation has at least one solution ,

called the trivial solution.

For a system of linear equations such as (2.1), we are interested in the following questions.

 Does the solution exist for the system?

 If the solution exists, how many solutions exist?
 Is there an efficient algorithm that computes actual solution?
Such a system of equations can be solved either by substitution or elimination or by matrix
method. Here we will use the matrix method. Matrices are a compact and convenient ways of
writing down system of linear equations.

1.2.9 Gauss elimination method

Definition: In a system of linear equation (2.1) or in its matrix notation (2.2) when we add on the
constant term matrix to the right hand of the coefficient matrix we obtain the
matrix

( | ,

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 15

 known as the augmented matrix of the system (2.1) or (2.2).

For example, for a system of linear equations

the augmented matrix corresponding to the system of linear equations is:

( | +

Theorem: Let and be two systems of linear equations with -equations and -
unknowns each. If the augmented matrix and are row equivalent, then the
original linear systems of equations are also equivalent. Therefore, they have exactly the
same solution.

Theorem: The non-homogenous system of equations is consistent if rank of the coefficient matrix
is equal to the rank of the augmented matrix. Otherwise it is inconsistent.

To solve a system of -equations with -unknowns is given by using the Gauss-Jordan elimination
method we use the following procedures.

Step-1: Find the augmented matrix.

Step-2: Use elementary row operation on the augmented matrix and transform the system into
row echelon form of a matrix.

Step-3: At last; use the back substitution method to solve for the unknowns.

Example: Solve the system of linear equation by augmentation method.

The corresponding augmented matrix is

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 16

 ( | + ( | + ( | +

( | +

The last matrix is in the row echelon form, hence the corresponding equivalent system of equation
is:

From equation , we observe that and put this in equation , then

Finally put the values of and in equation

Therefore, .

In solving system of linear equations (2.1) or (2.2) we start with the augmented matrix and
transform it into the row-echelon form. In such cases we have the following three alternatives.

Case-1 ( | , , we observe that rank of and rank of are equal and

the system of linear equations (1.1) or (1.2) has unique solution.

Case-2 ( | , , from the last row ( or some rows) we have which imply

that rank of and rank of are equal but less than the number of unknowns , then
the system of linear equations (1.1) or (1.2) has an infinitely many solution.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 17

Case-3 ( | , , from the last row we have (where is nonzero),

which implies that rank of is less than the rank of and this implies that the system
of linear equations (1.1) or (1.2) has no solution.

Example: Solve the following system of linear equation.

Solution: The augmented matrix will be:

⁄ ⁄ ⁄ ⁄
( | + ⁄ ( ⁄ ⁄ | ⁄ + ⁄
⁄ ⁄ ⁄ ⁄
⁄ ⁄ ⁄
( ⁄ | ⁄ +

The corresponding system of equation is: ,

We have two types of solutions:

1. The general solution of the system is

{ }

2. The particular solution:

If

If

If etc.

Example: Solve the following system of linear equation.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 18

 {

Solution: The augmented matrix will be:

⁄ ⁄ ⁄
( | + ( | + ⁄

( | +

The system has no solution as that is

Theorem: A homogenous system of -linear equations with -unknowns has a non-trivia solution
if and only if rank .

Example: Solve the following system of linear equation.

Solution: The augmented matrix will be:

⁄ ⁄ ⁄
( | + ⁄ ( ⁄ ⁄ ⁄ | + ⁄
⁄ ⁄ ⁄ ⁄ ⁄

⁄ ⁄ ⁄ ⁄
( ⁄ ⁄ | + ( ⁄ ⁄ | +
⁄ ⁄ ⁄ ⁄

Rank number of variables, hence we have non-trivial solutions. The corresponding system
of linear equations is:

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 19

 {

The general solution of the system is

{ }

And some particular solutions are:

If

If etc.

1.3 Determinant of a matrix

A determinant is a function that assigns a real number to each square matrix. Hence we will
consider how to find the determinant of square matrix of order .

A square matrix of order one is just the number system we have; thus

The determinant of a square matrix of order one is the number itself. That is if , then
.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 20

1.3.1 Determinant of matrix

For square matrix of order two ( ), its determinant denoted by or and is

given by:

| |

For example for the matrix ( ), its determinant is given by

| |

Definition: Let be matrix. If we delete the row and column of we obtain a sub-
matrix of dimension corresponding to the entry of matrix .

The number:

is called the minor of corresponding to the entry of matrix . is even

The number:

is called the cofactor of corresponding to the entry of matrix .

Example: Given the square matrix ( +, then if we delete the first row and the

second column, we have the sub-matrix ( ) with the corresponding minor will be

| | and cofactor corresponding to the element in the first row and the

second column of matrix .

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 21

Similarly if we delete the third row and the first column we have the sub-matrix ( )

with corresponding minor | | and cofactor corresponding to the element in

the third row and the first column of matrix is .

Now we can generalize the determinant of a matrix for any order based on the following
procedure on square matrix of order two.

For square matrix of order two ( ),

| |

; in terms of the minor as and

; in terms of the co-factor as and

1.3.2 Determinant of matrix

For matrix of order three ( +, then its determinant is given by:

| |

(in terms of the co-factor)

(In terms of the minor)

| | | | | |

In general for square matrix of order , * +, then

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 22

 | |

∑ , called expansion by the first row.

Example Find the determinant of the matrix ( +?

Solution: Using the expansion by the first row we have:

| | | |— | | | |

[ ] [ ] [ ]

[ ] [ ] [ ]

Example: Find the determinant of the matrix ( )

Solution: To find the determinant we use expansion by the first row

| | | | | | | |

{ | | | | | |} { | | | | | |}

{ | | | | | |} { | | | | | |}

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 23

 { } { } { }

{ }

Property of determinant of a matrix

Given a real number , and two square matrices of order , then we have the following
properties.

i)
ii)
iii)
iv) If two rows (or columns) are the same, then its determinant of the matrix is zero.
v) If two adjacent columns of are interchanged, then the determinant change in sign.
vi) If we add scalar multiple of a row (or column) vector to another row (or column) vector,
then the determinant does not change.
vii) If a row (or column) can be expressed as sum of two or more rows (or columns) then
the determinant is zero

Adjoint of a square matrix

Definition: Given a square matrix of order , the adjoint of , denoted by , is a square

matrix of order obtained by transposing the cofactor matrix of matrix . That is

( , ( ,

Example: Find the cofactor matrix of matrix ( ).

Solution: The sub-metrics of matrix are , and

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 24

thus the corresponding minors are , and , respectively (as
the determinant of a number is the number itself).The cofactor matrix corresponding to matrix

will be ( * ( ) and the transpose of the cofactor matrix of is called the adjoint

of and denoted by

Hence the adjoint of will be

( * ( ) ( )

1.3.3 Inverse of a matrix using determinant

Theorem: A square matrix of order is said to be non-singular or invertible if and only if

and hence the inverse of is given by

Example: If ( ), then find .

Solution: By using the above theorem, first find the determinant and then the adjoint of the .

| | , thus has an inverse.

To find the adjoint of : , and .

The adjoint of matrix will be the transpose of the cofactor matrix

( * ( ) ( )

Therefore, the inverse of matrix A will be:

⁄ ⁄
( ) ( *.
⁄ ⁄

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 25

Note: In general, for square matrix of order two ( ) and if , then the

inverse of is given by : ( )

Example: Find the inverse of the matrix ( +.

Solution: The determinant of :

| | | | | | | |

[ ] [ ] [ ]

To find the :

| | | | | |

| | | | | |

| | | | | |

Hence the adjoint of will be: ( + ( +

Therefore the inverse of will be:

⁄ ⁄ ⁄
( + ( ⁄ ⁄ ⁄ +
⁄ ⁄ ⁄

1.3.4 Cramer’s rule

Here we will use one of the matrix methods of solving system of equations when the number of
equation and the number of variables are equal.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 26

For a system of - linear equations with -unknowns is given by

One can use the square coefficient matrix, variable matrix, and the constant term matrix and
express equation as

( ,( , ( ,

In equation (2.4) as matrix is a square matrix we can find the determinant of . Thus the
Cramer’s rule will be stated as follows:

1. If , then the system (1.3) or (1.4) has a unique solution.

2. If , then the unique solution is given by:

where is the matrix obtained from the coefficient matrix by replacing its column by
the constant term matrix . This rule is called the Cramer’s Rule.

1.3.5 Inverse method

If we have -linear equations with -unknowns given by system of equation or
with , then

(as )
(by associative property)

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 27

Example: Solve the following system of linear equation by Cramer’s rule

Solution: In matrix representation we can express the given system as

( +( + ( +

| | | | | | | | . (Observe that the three rows or

columns of matrix are linearly independent)

| | | | | | | | .

| | | | | | | | .

| | | | | | | |

By the Cramer’s rule:

Also one can find and solve for the unknown variable matrix .

To find the ;

| | | | | |

| | | | | |

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 28

 | | | | | |

Hence: ( + ( +

Therefore, the solution will be:

( +( + ( + ( +

That is, and ; the same as the Cramer’s rule.

Applied Mathematics I by Mekonnen Elifnew (Assistant Professor) Page 29