An Autonomous Institution Affiliated to VTU, Belagavi Approveby AICTE, New Delhi

Recognized by UGC under 2(f) & 12(B) Accredited by NBA & NAAC

LINEAR ALGEBRA

Introduction

The knowledge of the determinants and matrices is necessary in various areas

of mathematics and a wide variety of other subjects. Determinants owe to birth to the
method of solving system of linear equations. In the language of matrices the system
of linear equation can be written in very simple form. The other areas of mathematics
where matrices occur include probability and statistics etc. Matrices are also used in
Physics, Economics, Psychology, Genetics and electrical Engineering.

Linear Algebra comprises of the theory and applications of linear system of

equation, linear transformations and Eigen value problems. In linear algebra, we make
a systematic use of matrices.

Elementary transformation of a matrix

Elementary row and column transformation

The following operations on a matrix are called Elementary transformation

Type I Operation: Interchanging any two rows (or columns). This transformation is
indicated by Rij or Ri  Rj (or Cij or Ci  Cj ) if the ith and jth row are interchanged

Type II Operation: Multiplications of the all the elements of the row (or column) by
a non zero scalar quantity k. Multiplying all the elements of the ith row (or column)
by k  0 is denoted by k Ri or Ri k Ri (k Ci or Ci k Ci )
Type III Operation: adding to the elements of the row (or column) k times the
corresponding elements of another row (or column).

Adding to the elements of the ith row (or column) k times the corresponding elements
of the jth row (or column) is denoted by Ri Ri + k Rj

The above operations are called elementary operations of elementary

transformations. An elementary transformation applied to only rows is called a row
elementary transformations and an elementary transformation applied to only columns
is called a column elementary transformations.

Examples:

−
A=
−

Then the row operation R2  R3 is applied to A yields the matrix

−
B= −

The row operation R2  3R2 is applied to B yields the matrix

−
C= −

The row operation R2  R1 + R2 is applied to C yields the matrix

−
D= −

Thus the matrix D is obtained from A by the successive elementary transformation.

Equivalent matrices

Two matrices of same order are said to be equivalent of one is obtained from
the other by elementary transformation. The symbol  is used to be equivalence.

Elementary matrix

A matrix obtained from a unit matrix by a single elementary transformation is

called a elementary matrix.

Example

0 1 0
1. I = 1 0 0 is an elementary matrix, obtained from I3 by interchanging
0 0 1
rows 1 and 2
1 0 0
2. 0 1 0 is an elementary matrix formed by adding 4 times row 1 of I3
−4 0 1
to row 3.

Note: All the elementary matrices are non-singular. Therefore each elementary
matrix possesses inverse.

Echelon Form

This lesson introduces the concept of an echelon matrix. Echelon matrices

come in two forms: the row echelon form (ref) and the reduced row echelon
form (rref).

Row Echelon Form

A matrix is in row echelon form (ref) when it satisfies the following conditions.

 The first non-zero element in each row, called the leading entry, is 1.
  Each leading entry is in a column to the right of the leading entry in the
previous row.
 Rows with all zero elements, if any, are below rows having a non-zero element.

Each of the matrices shown below are examples of matrices in row echelon form.

1 2 3 4
1 2 3 4 1 2
0 0 1 3
0 0 1 3 0 1
0 0 0 1
0 0 0 1 0 0
0 0 0 0
Aref Bref Cref

Note: Some references present a slightly different description of the row echelon
form. They do not require that the first non-zero entry in each row is equal to 1.

Reduced Row Echelon Form

A matrix is in reduced row echelon form (rref) when it satisfies the following
conditions.

 The matrix satisfies conditions for a row echelon form.

 The leading entry in each row is the only non-zero entry in its column.

Each of the matrices shown below are examples of matrices in reduced row echelon
form.
1 2 0 0
1 0 1 2 0 0
0 0 1 0
0 1 0 0 1 0
0 0 0 1
0 0 0 0 0 1
0 0 0 0

Aref Bref Cref

Normal forms or canonical form of a matrix:

The given matrix A is reduced to an Echlon form by first applying a series of

elementary row transformation.

Later column transformation are performed to reduce the matrix to one of the
following four forms called normal form.

I  I 0
1. Ir 2. [Ir, 0] 3.  r  4.  r 
0  0 0

Where Ir is the identity matrix of order r

Example
1 0 0 1 0 0 0 1 0  1 0 0 
1.  0 1 0
 
2.  0 1 0 0
   
3.  0 1  4.  0 1 0

0 0 1 0 0 1 0  0 0 0 0 0 
    

Rank of Matrix

2. The rank of a matrix A in its echelon form is equal to the number of non zero
rows. It is denoted by ( A)
3. The matrix A of order m x n can be reduced to one of the normal forms.
4. Therefore rank of the matrix ( A) = r

Problems

1. Find the rank of the following matrices by elementary row transformation

1 2 3 2
 
2 3 5 1
1 3 4 5 


Solution
 1 2 3 2
Let A =  2 3 5 1

1 3 4 5 


R2  R2-2R1, R3  R3 -R1
1 2 3 2 
A   0 1 1  3 
0 1 1 3 


R3  R3 + R2

1 2 3 2 
A   0 1 1  3 
0 0 0 0 


R2  R2 / (-1)

1 2 3 2
A   0 1 1 3 
0 0 0 0 
 

Thus the matrix A is the row echelon form is having two non zero rows ( A) = 2.

2. Find the rank of the following matrices by elementary row transformation

0 1 3  1 
 
1 0 1 1 
3 1 0 2 
 
1 1 2 0 

Solution:

0 1 3  1 
 
1 0 1 1 
A=  R1 R2
3 1 0 2 
 
1 1 2 0 
 1 0 1 1 
 
0 1 3  1 
A= R3  R3 -3R1 ; R4  R4- R1
3 1 0 2 
 
1 1 2 0

1 0 1 1 
 
0 1 3  1 
A= R3  R3 -R2 ; R4  R4- R2
0 1 3  1 
 
0 1  3  1

1 0 1 1 
 
0 1 3  1
A= C3  C3 -C2 ; C4  C4- C1
0 0 0 0 
 
0 0 0 0 

1 0 0 0 
 
 0 1 3  1 
A= C3  C3 +3C2 ; C4  C4+C2
0 0 0 0 
 
0 0 0 0 

1 0 0 0 
 
 0 1 0 0  Ir 0
A=   
0 0 0 0  0 0
 
0 0 0 0

 ( A) = 2.

Homework:
Find rank of matrix by reducing the matrix in normal form

 2 3 1  1  1 2 1 4 
   
 1 1 2  4   2 4 3 4 
1. A = 2. A =
3 1 3 2  1 2 3 4 
   
6 3 0 7   1  2 6  7 

Simultaneous Linear Equations

This lesson explains how to use matrix methods to (1) represent a system of linear
equations compactly and (2) solve simultaneous linear equations efficiently.
How to Represent a System of Linear Equations In Matrix Form

Suppose you have n linear equations with n unknowns. Using ordinary algebra, those
equations might be expressed as:

A11x1 + A12x2 + A13x3 + . . . + A1nxn = B1

A21x1 + A22x2 + A23x3 + . . . + A2nxn = B2
A31x1 + A32x2 + A33x3 + . . . + A3nxn = B3
...
An1x1 + An2x2 + An3x3 + . . . + Annxn = Bn

where

xj is an unknown value
Aij is the known coefficient of xj in equation i
Bj is a known quantity in equation j

This set of equations can be expressed compactly in matrix form as follows:

Ax = y

where

x is an n x 1 column vector of unknown values x1, x2, . . . , xn

A is an n x n matrix of the known coefficients Aij
B is an n x 1 column vector of known values B1, B2, . . . , Bn

GAUSS ELIMINATION METHOD

This method is illustrated by considering a system of three independent

equations in three unknowns
Consider the system of Equations

A11X1 + A12 X2 + A13X3 = B1

A21 X1 + A22 X2 + A23 X3 = B2 (1)
A31 X1 + A32 X2 + A33 X3 = B3

The system (1) is equivalent to the matrix Equation

AX = B (2)

A11 A12 A13 X1 B1

Where A= , X= , B=
A21 A22 A23 X2 B2

A31 A32 A33 X3 B3

The method aims in reducing the coefficient Matrix A to an upper triangular matrix.

We consider the augmented matrix [ A:B]

A11 ∶ 1
We have [ A: B ] = 21 22 23 : 2
31 32 33 ∶ 3

Step 1 : We use the element A11 ( 0) to make the elements A21 and A31 zero by
elementary row transformations

This transforms [A:B] into the form

11 1 1 ∶ 1
¢ ¢ ¢
[A:B] = 0 22 23 : 2 (3)
¢ ¢ ¢
0 32 33 ∶ 3

¢ ¢
Step 2 : we use the element 22 ( 0) to make element 32 by elementary row
transformations
 11 11 11 ∶ 1
¢ ¢ ¢
[A:B] = 0 22 23 : 2 (4)
² ¢¢
0 0 33 ∶ 3

( Observed that the matrix is reduced to upper triangular form )

From (4), the given system of linear equation is equivalent to the systems of equations

A11X1 + A12 X2 + A13X3 = B1

¢ ¢ ¢
22 X2 + 23 X3 = 2
² ¢¢
33 X3 = 3

We get X3 from the last of the equation and back substitutions we get X2 and X1 . the
values of X1 , X2 and X3 so obtained constitutes the exact solution of the given system
of equations.

¢
Note : In case 11 or 22 is zero we only need to rearrange the equations.

Problems
1. Solve the system of equations by Gauss elimination method
+ + =
− + =
+ − =

Soln:
The augmented matrix of the system is
1 1 1 :9 
 
[A:B] =  1 2 3 : 8 
 2 1 1 : 3 
 
R2  -R1 +R2, R3  -2R1 + R2

1 1 1 :9 
 
[A:B]   0 3 2 : 1 
 0 1 3 : 15 
 
R3R2 + (-3) R3
 1 1 1 :9 
 
[A:B]   0 3 2 : 1 
 0 0 11 : 44 
 
Hence we have,
+ + =9
−3 + 2 = −1
11 = 44
44
Þ = =4
11
 =4
By Back substitution,
−3 + 8 = −1  =3
Also,
+3+4= 9
 =2
Thus = , = , = is the required solution

2. Solve the system of equations by Gauss elimination method

+ + =
+ − =
− + =
Soln:
The augmented matrix of the system is
 2 1 4 :12 
 
[A:B] =  4 11 1 : 33 
 8 3 2 : 20 
 
R2  -2R1 +R2, R3  -4R1 + R3

2 1 4 :12 
 
[A:B]   0 9 9 : 9 
 0 7 14 : 28 
 
1/9. R2 and – 1/7. R3
  2 1 4 :12 
 
[A:B]   0 1 1 :1 
0 1 2 :4 
 
R3  -R2 +R3,

 2 1 4 :12 
 
[A:B]   0 1 1 :1 
0 0 3 :3 
 
Hence we have
2 + + 4 = 12
− =1
3 =3
 =1
By back Substitution, y = 2 and x = 3
Thus x = 3, y = 2 and z = 1 is required solution.

Class work:
1. Solve by Gauss elimination method
x1 – 2x2 + 3x3 = 2
3x1 – x2 + 4x3 = 4
2x1 +x2 - 2x3 = 5

Homework:
1. Apply the Gauss elimination method to solve system of Equations
+ + =
− + =
− + =
Eigen values and Eigen Vectors of Square Matrix
Definition:
Given a square matrix A, if there exists a scalar  ( real or complex) and
a non zero column matrix X such that AX = X, Then  is called an Eigen values of
A and X is called an Eigen vector of A corresponding to an Eigen value .
If I is a unit matrix of the same order as that of A, we have X = I X and hence AX =
X can be written as,
AX =  (IX) = (I) X
i.e., [A-I][X]=[0], here [0] is null matrix.

On expanding, we get the equation in  which is called the characteristic equation of

A. The roots of the equation are the Eigen value which also represents Eigen roots or
characteristic roots or latent roots. For each value of  there will be an Eigen vector
X  0 which is also called characteristic vector.

Problems:
1. Find the Eigen value and Eigen Vector of the following matrix.
 8 6 2 
A =  6 7 4 
 2 4 3 
 
Solution:
The characteristic equation of A is | − I| = 0
(8   )  6 2
i.e. 6 (7   )  4 0
2  4 (3   )

on expanding we have,
(8-) [(7-)(3-)-16]+6[-6(3-)+8]+2[24-2(7-)] = 0
i.e., (8-) [ 5-10+2]+6[6-10]+2[10+2] = 0
-3 - 182-45 = 0
On simplification
3 + 182+ 45 = 0
 (2 - 18 + 45) = 0 (or)  (-3)( -15) = 0
Hence  = 0, 3, 15 are the Eigen values of A.
We now form the system of equations
(8-)x -6y +2z = 0
-6x+(7-)y – 4z = 0
2x – 4y +(3-)z = 0 ...............................(1)

Case 1:let  = 0, the system of Equation (1) becomes,

8x -6y +2z = 0 .............................................(i)
-6x+7y – 4z = 0 ............................................ (ii)
2x – 4y +3z = 0 ..........................................(iii)
Apply the cross multiplication of Equation (i) and (ii)

x y z
 
6 2 8 2 8 6
7 4 6  4 6 7

i.e., = = ( ) = =

 (x, y, z) are proportional to (1, 2, 2) and we can write x= k, y = 2k , z = 2k

where k is arbitrary. However it is enough to keep the values of (x, y, z) in the
simplest form x= 1, y = 2, z = 2. These values satisfy all the equations
simultaneously.
Thus the Eigen vector of the corresponding Eigen value  = 0 is
1 
X 1   2
 2

[ the same will be written as a row vector in the form X1 = [1 2 2] in future]

Case 2:let  = 3, the system of Equation (1) becomes,
5x -6y +2z = 0 .............................................(iv)
-6x+4y – 4z = 0 ............................................ (v)
 2x – 4y +0z = 0 ..........................................(vi)
Apply the cross multiplication of Equation (iv) and (v)

x y z
 
24  8 20  12 20  36

i.e., = = ( ) = =


X2 = [2, 1, -2] is the eigen vector of the corresponding Eigen value  = 3

Case 3:let  = 15, the system of Equation (1) becomes,

-7x -6y +2z = 0 .............................................(vii)
-6x-8y – 4z = 0 ............................................ (viii)
2x – 4y -12z = 0 ..........................................(ix)
Apply the cross multiplication of Equation (vii) and (viii)

x y z
 
24  16 28  12 56  36

i.e., = = ( ) = =


X3 = [2, 2, 1] is the eigen vector of the corresponding Eigen value  = 15.

2. Find the Eigen value and Eigen Vector of the following matrix.
 6 2 2 
A =  2 3 1
 2 1 3 
 
Solution:
The characteristic equation of A is | − I| = 0
(6   )  2 2
i.e. 2 (3   )  1 0
2  1 (3   )
On expanding we have,
3 + 182+ 45 = 0
 = 2 is a root by inspection, now by synthetic division method
1  12 36  32
2 0  2  20  32
1  10 16 0

 Now by solving
2 - 10 + 16 = 0 we obtain ( − 2)( − 8) = 0

  = 2, 8

Hence  = 2,2, 8
We now form the system of equations
(6-)x -2y +2z = 0
-2x+(3-)y –1z = 0 ...............................(1)
2x – 1y +(3-)z = 0
Case 1:let  = 2, the system of Equation (1) becomes,
4x -2y +2z = 0 .............................................(i)
-2x+y – z = 0 ............................................ (ii)
2x – y +z = 0 ..........................................(iii)
The above set of equation are all same as we have only one dependent equation
2x – y +z = 0 and hence we can choose two variables arbitrarily.
Let z =k1 , y = k2

 x = (k2 – k1) / 2

 X1 = [(k2 – k1) / 2, k2, k1] is the Eigen vector corresponding to  = 2 where

k1 and k2 are not simultaneously equal to zero.

Case 2:let  = 8, the system of Equation (1) becomes,
2x -2y +2z = 0 .............................................(iv)
-2-5y – z = 0 ............................................ (v)
 2x – y-5z = 0 ..........................................(vi)

Þ i.e., = = ( ) = =

X2 = [2, -1, 1] is the eigen vector of the corresponding Eigen value  = 8

Homework:
 7 2 0 
1. A =  2 6 2 
 0 2 5 
 

 2 2 3 
2. A =  2 1 6 
 1 2 0 
 

***********************