Calculus & Linear Algebra (22MAT11)

BANGALORE INSTITUTE OF TECHNOLOGY

Department of Mathematics
Module 5

Definition:
A set of mn elements (real or complex) written in rectangular array of ' m ' rows and ' n '
columns is called as m  n matrix and m  n matrix can be written as
 a 11 a 12 a13 a 1n 
a a a a2 4 
 21 2 2 2 3 
 
a a a m n 
 m1 m 2 am 3

Elementary operations:
Let ' A ' be the given matrix, then the following are called row operation on ' A '
i) Interchange of any two rows i.e., R i  R j .
ii) Multiplication of the elements of a row by a non-zero constant i.e., R i  k R i .
iii) Addition of the elements of a row with k times the corresponding elements of
any other row i.e., R i  R i  k R j .

Equivalent matrices:
Two matrices ' A ' and ' B ' of the same order are said to be equivalent if one of the matrix
can be obtained from the other by successive elementary operations. It is denoted by
A B.

Reduction of a matrix to row reduced echelon form:

A matrix of order m  n is said to be row reduced echelon form if
i) The leading element (the first non-zero entry) of each row is one.
ii) All the entries below the leading entry are zero.

DEPARTMENT OF MATHEMATICS, B.I.T

1
 Calculus & Linear Algebra (22MAT11)

iii) The zero rows must appear below the non-zero row.
iv) The number of zeroes appearing before the leading entry in each row is greater
than that appears in its previous row.

Rank of a matrix:
Let ' A ' be a non-zero matrix then a positive integer ' r ' is said to be a rank of ' A ' if the
following conditions are satisfied.
i) There exist atleast one minor of order ' r ' of matrix ' A ' which doesn’t vanish.
ii) All minors of the order  r  1 must vanish.
(Or)
The rank of a matrix ' A ' is equal to number of non-zero rows in the matrix. It is denoted
by   A  .

Remark:
i) If ' A ' is null matrix, then   A   0 .
ii) If ' A ' is not a null matrix, then   A   1.
iii) If ' A ' is matrix of order m  n then   A   min  m, n  .
iv) If ' A ' is a square matrix of order ' n ' then   A   n .
v) For a singular matrix ' A ' ,   A   n .
vi) For a non-singular matrix ' A ' of order ' n ' ,   A   n .
vii) Rank and order remain same when elementary transformations are applied.

DEPARTMENT OF MATHEMATICS, B.I.T

2
 Calculus & Linear Algebra (22MAT11)

Problems:
Using elementary transformations reduce the following matrix to the echelon form and
also find the rank of the matrix
1 3 2 
1.  2 1 4 
 
1 11 14 

1 3 2 
Solution: Let A =  2 1 4 
 
1 11 14 

1 3 2 
R2  R2  2 R1 0  7 8
R3  R3  R1  
1 14 16 

1 3 2 
R3  R3  2R2 0 7 8 
 
1 0 0 

   A  2

4 0 2 1
2 1 3 4
2.  
2 3 4 7
 
2 3 1 4

DEPARTMENT OF MATHEMATICS, B.I.T

3
 Calculus & Linear Algebra (22MAT11)

4 0 2 1
2 1 3 4
Solution: Let A =  
2 3 4 7
 
2 3 1 4
2 1 3 4
4 0 2 1
R1  R2  
2 3 4 7
 
2 3 1 4
2 1 3 4
R2  R2  2 R1  0 2 4 7 
R3  R3  R1  
0 2 1 3
R4  R4  R1  
 0 2 2 0 
2 1 3 4
R3  R3  R2  0 2 4 7
 
R4  R4  R2  0 0 3  4
 
 0 0 6  7

2 1 3 4
 0 2 4  7
R4  R4  2R3  
 0 0 3  4
 
0 0 0 0

   A  4

 2 3 1 1
1 1 2 4 
3.  
3 1 3 2 
 
6 3 0 7 

DEPARTMENT OF MATHEMATICS, B.I.T

4
 Calculus & Linear Algebra (22MAT11)

 2 3 1 1
1 1 2 4 
Solution: Let A =  
3 1 3 2 
 
6 3 0 7 
1 1 2 4 
 2 3 1 1
R1  R2  
3 1 3 2 
 
 6 3 0 7 
1 1 2 4 
R2  R2  2 R1 0 5 3 7 
R3  R3  3R1  
0 4 9 10 
R4  R4  6 R1  
0 9 12 17 
4 1 1 2 4 
R3  R3   R2  0 5 3 7 
5  
9 0 0 33/ 5 22 / 5
R4  R4   R2   
5 0 0 33/ 5 22 / 5
1 1 2 4 
0 5 3 7 
R4  R4  R3  
0 0 33/ 5 22 / 5 
 
0 0 0 0 

   A  3

 2 3 1 1
1 1 2 4 
4.  
3 1 3 2 
 
6 3 0 7 

DEPARTMENT OF MATHEMATICS, B.I.T

5
 Calculus & Linear Algebra (22MAT11)

 2 3 1 1
1 1 2 4 
Solution: Let A =  
3 1 3 2 
 
6 3 0 7 
1 1 2 4 
 2 3 1 1
R1  R2  
3 1 3 2 
 
 6 3 0 7 
1 1 2 4 
R2  R2  2 R1 0 5 3 7 
R3  R3  3R1  
0 4 9 10 
R4  R4  6 R1  
0 9 12 17 
4 1 1 2 4 
R3  R3   R2  0 5 3 7 
5  
9 0 0 33/ 5 22 / 5
R4  R4   R2   
5 0 0 33/ 5 22 / 5
1 1 2 4 
0 5 3 7 
R4  R4  R3  
0 0 33/ 5 22 / 5 
 
0 0 0 0 

   A  3

0 1 3 1
1 0 1 1
5.  
3 1 0 2
 
1 1 2 0 

DEPARTMENT OF MATHEMATICS, B.I.T

6
 Calculus & Linear Algebra (22MAT11)

0 1 3 1
1 0 1 1
Solution: Let A =  
3 1 0 2
 
1 1 2 0 
1 0 1
1
0 1 3 1
R1  R2  
3 1 0 2
 
1 1 2 0 
1 0 1 1
R3  R3  3R1 0 1 3 1 
 
R4  R4  R1 0 1 3  1
 
0 1 3  1
1 0 1 1
R3  R3  R2 0 1 3 1
 
R4  R4  R2 0 0 0 0
 
0 0 0 0

   A  2

 91 92 93 94 95 
92 93 94 95 96 
 
6. 93 94 95 96 97 
 
94 95 96 97 98 
95 96 97 98 99 

DEPARTMENT OF MATHEMATICS, B.I.T

7
 Calculus & Linear Algebra (22MAT11)

 91 92 93 94 95 
92 93 94 95 96 
 
Solution: Let A = 93 94 95 96 97 
 
94 95 96 97 98 
95 96 97 98 99 
91 92 93 94 95
R5  R5  R4 1 1 1 1 1
R4  R4  R3  
1 1 1 1 1
R3  R3  R2  
R2  R2  R1 1 1 1 1 1
 1 1 1 1 1 

91 92 93 94 95
R5  R5  R2 1 1 1 1 1
 
R4  R4  R2 0 0 0 0 0
 
R3  R3  R2  0 0 0 0 0 
 0 0 0 0 0 

1 1 1 1 1 
91 92 93 94 95 
 
R1  R2 0 0 0 0 0
 
 0 0 0 0 0 
 0 0 0 0 0 

1 1 1 1 1
0 2 3 4 5
 
R2  R2  91R1 0 0 0 0 0
 
0 0 0 0 0
0 0 0 0 0 

   A  2

DEPARTMENT OF MATHEMATICS, B.I.T

8
 Calculus & Linear Algebra (22MAT11)

Solution of system of linear equations:

Let us consider system of ' m ' simultaneous linear algebraic equations in ' n ' unknowns
i.e.,
a 11 x1  a 12 x2  a 13 x3  ........a 1n xn  b1
a 21 x1  a 22 x2  a 23 x3  ........a 2 n xn  b2 
 1
..................................................  ... 
am1 x1  am 2 x2  am 3 x3  ........am n xn  bm 

Here x1 , x2 , x3 ......... xn are unknowns and ' a ' s and ' b ' s are known as real constants.

Homogenous system: if ' b ' s are zero in 1 .

Non-homogenous system: if at least one of the ' b ' s is not zero in 1 .

Solution for homogeneous system of linear equations:

Solution: A set of values of x1 , x2 , x3 ......... xn that satisfies all the equation of the system 1
is called solution.
If x1  x2  x3    xn  0 satisfies the system of equation then the solution is called
Trivial solution. A solution which is not trivial solution is called non-trivial solution.

NOTE
n  number of variables
m  number of equations
r  rank of the matrix

DEPARTMENT OF MATHEMATICS, B.I.T

9
 Calculus & Linear Algebra (22MAT11)

Worjking rule to solve Homogeneous system of linear equations

1. Consider the co-efficient matrix A from the given system of equation.

2. Reduce the matrix to echelon form and find the rank of the matrix.
3. If r  n then the system has trivial solution.
4. If r  n then the system has a non-trivial solution.
i) Consider the equation corresponding to the matrix in Echelon form.
ii) Consider r  n variables as constant.
iii) Find the value of the variable in terms of constants.

Problems:
Solve the following system of equations.
1. x  2 y  3z  0
3x  4 y  4 z  0
7 x  10 y  12 z  0

1 2 3 
Solution: A  3 4 4 
 
7 10 12
1 2 3
R2  R2  3R1 
~ 0  2  5
R3  R3  7 R1  
0  4  9
1 2 3
R3  R3  2R2 ~ 0  2  5

0 0 1 
  A  3  r
Number of variables  n  3
r  n  The system has trivial solution
x  y  z  0

2. 2 x  y  3z  0

DEPARTMENT OF MATHEMATICS, B.I.T

10
 Calculus & Linear Algebra (22MAT11)

3x  2 y  z  0
x  4 y  5z  0
2  1 3
Solution: A   3 2 1
 
1  4 5
1  4 5
R1  R3 ~ 3 2 1
 
2  1 3
1 4 5 
R2  R2  3R1 
~ 0 14 14 
R3  R3  2 R1  
0 7 7 
R2  2
R 1 4 5 
14 
~ 0 1 1
R  
R3  3 
0 1 1
7
1 4 5 
R3  R3  R2 ~ 0 1 1
0 0 0 
  A  2  r
Number of variables  n  3
 r  n  The system has non-trivial solution
Consider the equations based on matrix in Echelon form
x  4 y  5z  0 1
yz0  2
Consider n  r  3  2  1variableasconstant.
If z  k
 2  y  k  0  y  k
1  x  4k  5k  0  x  k
x  k , y  k , z  k

DEPARTMENT OF MATHEMATICS, B.I.T

11
 Calculus & Linear Algebra (22MAT11)

Non-homogeneous system of linear equations

a 11 x1  a 12 x2  a 13 x3  ........a 1n xn   b1
a 21 x1  a 22 x2  a 23 x3  ........a 2 n xn  b2 
 1
..................................................  ... 
am1 x1  am 2 x2  am 3 x3  ........am n xn  bm 

Matrix form A X  B .

 a 11 a 12 a 13 .... a 1n   x1  b 1 
a a 22 a 23 .... a 2 n  x  b 
 21   2 = 2 
 .... .... .... .... ....   ...   ... 
     
 am 1 a m 2 am 3 .... am n   xn  bm 

 a 11 a 12 a 13 .... a 1n 
a a 22 a 23 .... a 2 n 
where A    is m  n matrix called as co-efficient matrix
21

 .... .... .... .... .... 

 
 am 1 am 2 am 3 .... am n 

 x1 
x 
X   2 is a column vector of order n  1
 ... 
 
 xn 

b 1 
b 
B   2  is a column vector of order m  1
 ... 
 
bm 

Augumented Matrix: Suppose the given equation is non homogenous than the co-
efficient matrix ' A ' and extra column consisting of the element ' B ' can be written as

DEPARTMENT OF MATHEMATICS, B.I.T

12
 Calculus & Linear Algebra (22MAT11)

 a 11 a 12 a 13 .... a 1n : b1 
a a 22 a 23 .... a 2 n : b2 
 A : B    .... .... .... .... .... : ... 
 21

 
 am1 am 2 am 3 .... am n : bm 
This is m   n  1 matrix is called augumented matrix. This is also denoted by
 A , B  or  A / B  .
Consistency & Inconsistency: The system of equation is said to be consistent if it
possess a solution and inconsistent if it doesn’t possess a solution. 1

Working rule:

1. Consider the augumented matrix  A : B  from given system of equations.

2. Reduce the augumented matrix  A : B  to the echelon form by using appropriate
elementary row operations.
3. Find the rank of ' A ' and  A : B  .
i) If the ranks are equal than the system is consistent.
ii) If the ranks are not equal than the system is inconsistent.
4. If the system is consistent write down the reduced form of linear equation.
If r  n, then n  r unknowns in the reduced system can be chosen arbitrarily,
which gives infinite solution.
If r  n, then the reduced system will have unique solution.

Problems:
1) Show that the system of equations
x  y  z  3 ; 3x  y  2 z  2 ; 2 x  4 y  7 z  7 is not consistent

Solution: Given x  y  z  3 ; 3x  y  2 z  2 ; 2 x  4 y  7 z  7
1 1 1  3  x
Let A   3 1 2  ; B   2  & X   y
     
 2 4 7   7   z 

DEPARTMENT OF MATHEMATICS, B.I.T

13
 Calculus & Linear Algebra (22MAT11)

1 1 : 3 
1
 A : B   3 1  2 : 2 

 2 4 7 : 7 

1 1 1 : 3 
R2  R2  3R1 0 2 5 : 7
R3  R3  2 R1  
0 2 5 : 13

1 1 1 : 3 
R3  R3  R2 0 2 5 : 7
 
0 0 0 : 20 

   A  2&   A : B   3

  A    A : B  The system is inconsistent

2) Show that the following system of equations does not possess any solution
5 x  3 y  7 z  5 ; 3x  26 y  2 z  9 ; 7 x  2 y  10 z  5
Solution: Given 5 x  3 y  7 z  5 ; 3x  26 y  2 z  9 ; 7 x  2 y  10 z  5
5 3 5 4  x
Let A   3 26 2 ; B  9  & X   y
     
7 2 10   5   z 
5 37 : 5
 A : B   3 26 2 : 9 

7 2 10 : 5 

3 5 5 
R2  R2    R1  3 7 :
5 0 121  11 : 30 
7  5 5 5 
R3  R3    R1 0
5   11 1 : 10 
5 5 5

DEPARTMENT OF MATHEMATICS, B.I.T

14
 Calculus & Linear Algebra (22MAT11)

5 3 7 4 :
R2  5R2 0 121  11 : 30 
R3  5R3  
0  11 1 : 10 

5 3 7 : 4 
R   
R3  R3   2  0 11 1 : 30 
 11   
0 0 0 : 80 
11
   A  2&   A : B   3

  A    A : B  The system is inconsistent

3) Discuss the consistency of system of equations and hence

2 x  3 y  4 z  11 ; x  5 y  7 z  15 ; 3x  11y  13z  25
Solution: Given 2 x  3 y  4 z  11 ; x  5 y  7 z  15 ; 3x  11y  13z  25
2 3 4 11   x
Let A  1 5 7 ; B  15  & X   y
     
 3 11 13  25  z 

2 3 4 : 11 
 A : B   1 5 7 : 15 

 3 11 13 : 25 

1 7 : 15 
5
R1  R2 2 3 4 : 11 
 
 3 11 13 : 25 

1 5 7 :15 
R2  R2  2 R1 0 7  10 : 19 
R3  R3  3R1  
0 4  8 : 20 

DEPARTMENT OF MATHEMATICS, B.I.T

15
 Calculus & Linear Algebra (22MAT11)

1 5 7 : 15 
R3  3
R 0 7  10 : 19 
 4   
0 1 2 : 5 

1 5 : 15 
7
0 1 2 : 5
  R2   R3  
0 7 10 : 19 

1 5 15 
7 :
R3  R3  7R2 0 1 2 : 5
 
0 0 4 : 16 

where n= number of unknowns    A  3,   A : B   3& n  3

  A    A : B   n The system is consistent with a unique solution.    n 

Let us write down the reduced form of system of equations, to find the unknowns
x  5 y  7 z  15 i 
y  2z  5  ii 
 4 z   16  z  4

Substituting in  ii 
 ii   y  2  4   5
 y  58
 y  3

Substituting y & z in  i 

i   x  5 y  7 z  15
 x  5  3  7  4   15
 x  15  15  28
 x2

DEPARTMENT OF MATHEMATICS, B.I.T

16
 Calculus & Linear Algebra (22MAT11)

 2
 X   3   X   2 3 4
T
 
Or
 4 

4) Show that the following system of equations is consistent and solve them
x  2 y  2 z  1 ; 2 x  y  z  2 ; 3x  2 y  2 z  3 ; y  z  0
Solution: Given x  2 y  2 z  1 ; 2 x  y  z  2 ; 3x  2 y  2 z  3 ; y  z  0
1 2 2 1 
2  2   x
 ; B    & X   y
1 1
Let A  
3 2 2 3   
     z 
0 1 1 0 
1 2 2 : 1
2 1 1 : 2
 A : B   
3 2 2 : 3
 
0 1 1 : 0
1 2 2 : 1
R2  R2  2 R1  0 3 3 : 0
 
R3  R3  3R1  0 4 4 : 0
 
0 1 1 : 0

R2   3
R2 1
0

2 2 : 1
1 1 : 0

0 1 1 : 0
R3   3 
R
 4   
0 1 1 : 0
 1 2 2 : 1
R3  R3  R2  0 1 1 : 0
 
R4  R4  R2  0 0 0 : 0
 
 0 0 0 : 0

DEPARTMENT OF MATHEMATICS, B.I.T

17
 Calculus & Linear Algebra (22MAT11)

where n= number of unknowns    A  2,   A : B   2 & n  3

  A    A : B  &   n The system is consistent    A    A : B   with an

infinite solution.
Let us write down the reduced form of system of equations, to find the unknowns
x  2 y  2z  1 i 
yz 0  ii 
  n i.e 3  2 , let us choose 3  2  1 arbitrary constant, i.e., let z  k

Substituting z  k in  ii 

 ii   y   k   0
 y  k
 y  k

Substituting y & z in  i 

i  x  2 y  2z  1
 x  2  k   2  k   1
 x 1

 1
 X   k  X  1 k k
T
  Or
 k 

5) Discuss the consistency of system of equations and hence

5 x  3 y  7 z  4 ; 3x  26 y  2 z  9 ; 7 x  2 y  10 z  5
Solution: Given 5 x  3 y  7 z  4 ; 3x  26 y  2 z  9 ; 7 x  2 y  10 z  5

DEPARTMENT OF MATHEMATICS, B.I.T

18
 Calculus & Linear Algebra (22MAT11)

5 3 7 4  x
Let A   3 26 2 ; B  9  & X   y
     
7 2 10   5   z 

5 3 7 : 4
 A : B   3 26 2 : 9 

7 2 10 : 5 

3 5 4 
R2  R2    R1  3 7 :
5 0 121  11 : 33 
7  5 5 5
R3  R3    R1 0
5   11 1 :  3 
5 5 5
5 5 3 7 : 4
R2    R2 0
 11  11 1 : 3 
 
R3  5 R3 0  11 1 : 3

5 3 7 : 4
R3  R2  R3 0 11  1 : 3
 
0 0 0 : 0 

where n= number of unknowns    A  2,   A : B   2 & n  3

  A    A : B  &   n The system is consistent    A    A : B   with an

infinite solution.
Let us write down the reduced form of system of equations, to find the unknowns
5x  3 y  7 z  4 i 
11y  z  3  ii 
  n i.e 3  2 , let us choose 3  2  1 arbitrary constant, i.e., let z  k

Substituting z  k in  ii 

DEPARTMENT OF MATHEMATICS, B.I.T

19
 Calculus & Linear Algebra (22MAT11)

 ii   11 y  z  3
11 y   k   3
 11 y  3  k
3 k
 y
11

Substituting y & z in  i 

i   5x  3 y  7 z  4
3 k 
 5x  3   7k  4
 11 
9 3k
 5x    7k  4
11 11
9 80k
 5x   4
11 11
9 80k
 5x  4  
11 11
9 80k
 5x  4  
11 11
35 80k
 5x  
11 11
35 80k
 x 
55 55

T
 35 80k   3  k  
 X       k
 55 55   11  

6) Find the values of  &  for which the system

x  y  z  6 ; x  2 y  3z  10 ; x  2 y   z  
a) Unique solution b) Infinite solution c) No solution
Solution: Given x  y  z  6 ; x  2 y  3z  10 ; x  2 y   z  

DEPARTMENT OF MATHEMATICS, B.I.T

20
 Calculus & Linear Algebra (22MAT11)

1 1 1 6
Let A  1 2 3 ; B  10 
   
1 2     

1 1 6
1 :
 A : B   1 2 3 : 10 

1 2  :  

1 1 1 6 
:
R2  R2  R1
 0 1 2 : 4 
R3  R3  R1  
0 1   1 :   6 

1 1 1 : 
6
R3  R3  R2  0 1 2 : 4 
 
0 0   3 :   10 
a)Unique solution
If   A    A : B  there is a solution and   n we will have unique solution.

Here if   A    A : B   3 and   n  3

If    3  0 or   3 irrespective of the value of  gives a unique solution.

The system will have unique solution if   3

b) Infinite solution
If   A    A : B  there is a solution and   n we will have infinite solution.

Here if   A    A : B   2 & n  3

If    3  0 &    10   0    3 &   10 gives an infinite solution.

The system will have infinite solution if   3 &   10

DEPARTMENT OF MATHEMATICS, B.I.T

21
 Calculus & Linear Algebra (22MAT11)

c)No solution
If   A    A : B  there is a no solution.

Here if   A  2 ;   A : B   3

If    3  0 &    10   0    3 &   10 gives no solution.

The system will have no solution if   3 &   10

7) Find for what values of k the system of equations

x  y  z  1 ; x  2 y  4 z  k ; x  2 y  10 z  k 2 posses a solution. Solve
completely in each case.

Solution: Given x  y  z  1 ; x  2 y  4 z  k ; x  2 y  10 z  k 2
1 1 1 1  x
Let A  1 2 4 ; Bk  ; X   y
     
1 4 10   k 2   z 

1 1 1
1 :
 A : B   1 2 4 : k

1 4 10 : k 
2

1 1 1 
1 :
R2  R2  R1 0 1 3 : k 1 
R3  R3  R1  
0 3 9 : k  1
2

1 1 1 :  1
R3  R3  3R2 0 1 3 : k 1  1
 
0 0 0 : k  3k  2 
2

DEPARTMENT OF MATHEMATICS, B.I.T

22
 Calculus & Linear Algebra (22MAT11)

Here   A   2 and for the system to be consistent it is must that   A : B   2 . And it is

possible if k 2  3k  2  0   k  1 k  2   0  k  1, k  2

 The system will possess a solution if k  1, k  2

   A  2,   A : B   2 & n  3

  A    A : B  &   n The system is consistent    A    A : B   with an

infinite solution.
Case1: when k  1 in 1

Let us write down the reduced form of system of equations, to find the unknowns

x  y  z 1 i 
y  3z  0  ii 
  n i.e 3  2 , let us choose 3  2  1 arbitrary constant, i.e., let z  k 1

Substituting z  k 1 in  ii 
 ii   y  3 k 1   0
 y   3k 1
 y   3k 1

Substituting y & z in  i 

i   x  y  z 1
 x   3k1    k1   1
 x  1  2k1

DEPARTMENT OF MATHEMATICS, B.I.T

23
 Calculus & Linear Algebra (22MAT11)

 1  2k1 
 X   3k1  X  1  2k1 3k1 k1 
T
 
Or
 k1 

Case2: when k  2 in 1

Let us write down the reduced form of system of equations, to find the unknowns

x  y  z 1 i 
y  3z  1  ii 
  n i.e 3  2 , let us choose 3  2  1 arbitrary constant, i.e., let z  k 2

Substituting z  k 2 in  ii 
 ii   y  3 k 2   1
 y  1  3k 2
 y  1  3k 2

Substituting y & z in  i 

i   x  y  z 1
 x  1  3k2    k2   1
 x  2k2

 2k2 
 X  1  3k2  X   2k2 1  3k2 k2 
T
  Or
 k2 

Gauss Elimination Method:

Gauss elimination method aims in reducing the coefficient matrix ' A ' to an upper
triangular matrix.

DEPARTMENT OF MATHEMATICS, B.I.T

24
 Calculus & Linear Algebra (22MAT11)

Note:
Gauss elimination method is very much similar to the method we employed in reducing a
system of equations by testing its consistency.
Problems:
1) Solve by Gauss elimination method
2 x  y  4 z  12 ; 4 x  11y  z  33 ; 8 x  3 y  2 z  20
Solution: Given 2 x  y  4 z  12 ; 4 x  11y  z  33 ; 8 x  3 y  2 z  20
2 1 4 12   x
Let A   4 11  1 ; B   33  & X   y
     
8 3 2   20   z 

2 1 4 : 12 
 A : B   4 11  1 : 33 

8 3 2 : 20 

2 1 4 12 :
R2  R2  2 R1 0 9 9 : 9 
R3  R3  4 R1  
 0 7  14 : 28
R2
R2  2 1 4 : 12 
9 0 1 1 : 1 
R  
R3  3  0 1 2 : 4 
 7 
2 4 : 12 
1
R3  R3  R2 0 1 1 : 1 
 
 0 0 3 : 3 
Let us write down the reduced form of system of equations, to find the unknowns
2 x  y  4 z  12 i 
y  z 1  ii 
3z  3  z 1  iii 

DEPARTMENT OF MATHEMATICS, B.I.T

25
 Calculus & Linear Algebra (22MAT11)

Substituting  iii  in  ii 
 ii   y  z  1
 y 1  1
 y2

Substituting y & z in  i 

i   2 x  y  4 z  12
 2 x  2  4 1  12
 2 x  12  2  4
 2x  6
 x3

3
 X  2 X  3 2 1
T
 
Or
1 

2) Solve by Gauss elimination method

x  4 y  z  5 ; x  y  6 z  12 ; 3x  y  z  4
Solution: Given x  4 y  z  5 ; x  y  6 z  12 ; 3x  y  z  4
1 4  1  5   x
Let A  1 1  6 ; B   12  & X   y
     
3 1  1   4   z 

1 4 1 : 5 
 A : B   1 1 6 : 12 
3 1 1 : 4 

DEPARTMENT OF MATHEMATICS, B.I.T

26
 Calculus & Linear Algebra (22MAT11)

1 4 1 : 5 
R2  R2  R1 0  3 5 : 12 
R3  R3  3R1  
0 13 2 : 4 

1 4 1 5  :
R3  R3  4R2 0 3 5 : 12 
 
0 1 22 : 47 

1 4 1 : 5 
0 47 
  R3     R2  
1 22 :

0 3 5 : 7 

1 4 1 : 5 
R3  R3  3R2 0 1 22 :  47 
 
0 0 71 : 148 
Let us write down the reduced form of system of equations, to find the unknowns
x  4 y  z  5 i 
y  22 z  47  ii 
148
71z  148  z  iii 
71
Substituting  iii  in  ii 
 ii   y  22 z  47
 148 
 y  22    47
 71 
81
 y
71

Substituting y & z in  i 

DEPARTMENT OF MATHEMATICS, B.I.T

27
 Calculus & Linear Algebra (22MAT11)

i   x  4 y  z  5
 81   148 
 x  4    5
 71   71 
117
 x
71

  117  
  71  
 
 81   81 148 
T
117
 X    X  
71  
Or
  71 71 71 
 
  148  
  71  
 

3) Solve by Gauss elimination method

x  2 y  z  3  0; 2 x  3 y  2 z  5 ; 3x  5 y  5 z  2  0
Solution: Given x  2 y  z  3  0; 2 x  3 y  2 z  5 ; 3x  5 y  5 z  2  0
1 2 1  3  x
Let A   2 3 2 ; B  5 & X   y
     
 3 5 5   2   z 

1 2 1 : 3 
 A : B   2 3 2 : 5 
 3 5 5 : 2 

1 2 1 : 3
R2  R2  2 R1 0 1 0 : 1
R3  R3  3R1  
0 11 2 : 7 

1 2 1 : 3
R3  R3  11R2 0 1 0 : 1
 
0 0 2 : 4 
Let us write down the reduced form of system of equations, to find the unknowns

DEPARTMENT OF MATHEMATICS, B.I.T

28
 Calculus & Linear Algebra (22MAT11)

x  2y  z  3 i 
 y  1  y 1  ii 
2z  4  z 2  iii 
Substituting y & z in  i 

i   x  2y  z  3
 x  2 1  2  3
 x  1

 1
 X   1  X   1 1 2
T
 
Or
 2 

4) Solve by Gauss elimination method

x  2 y  2 z  2 ; 3x  2 y  z  5 ; 2 x  5 y  3z  4 ; x  4 y  6 z  0
Solution: Given
x  2 y  2 z  2 ; 3x  2 y  z  5 ; 2 x  5 y  3z  4 ; x  4 y  6z  0
1 2 2   2
 3 2 1  5  x
Let A    ; B  & X   y
 2 5 3   4   
     z 
1 4 6   0
1 2 2 : 2
 3 2 1 : 5
 A : B   
 2 5 3 : 4 
 
1 4 6 : 0
1 2 2 : 2
R2  R2  3R1 0 8  7 : 1
R3  R3  2 R1  
0 9  1 : 8
R4  R4  R1  
0 2 4 : 2 

DEPARTMENT OF MATHEMATICS, B.I.T

29
 Calculus & Linear Algebra (22MAT11)

1 2 2 : 2
R2    R2  0 8 7 : 1
R3  R3  R2  
0 1 6 : 7 
R4 
R4  
0 1 2 : 1
2
1 2 2 : 2
0 1 6 : 7 
R2   R3   
0 8 7 : 1
 
0 1 2 : 1
1 2 2 : 2 
R3  R3  8  R2  0 1 6 : 7 
 
R3  R3  R2 0 0 55 :  55 
 
0 0 8 : 8 

R3 1 2 2 : 2
R3  0 1
55 6 : 7 
 
R 0 0 1 : 1
R4  4  
8 0 0 1 : 1
1 2 2 : 2
0 1 6 : 7 
R4  R4  R3  
0 0 1 : 1
 
0 0 0 : 0
Let us write down the reduced form of system of equations, to find the unknowns
x  2 y  2z  2 i 
 y  6 z  7  ii 
z  1  iii 
Substituting  iii  in  ii 

DEPARTMENT OF MATHEMATICS, B.I.T

30
 Calculus & Linear Algebra (22MAT11)

 ii    y  6 z  7
  y  6  1  7
 y 1

Substituting y & z in  i 

i   x  2 y  2z  2
 x  2 1  2  1  2
 x2

 2 
X   1   X   2 1 1
T
 
Or
 1 

Gauss Jordon Method:

Gauss Jordon method is the modification of Gauss elimination method. It aims in
reducing the coefficient matrix ' A ' to its diagonal matrix by elementary row operations.
Problems:
1) Solve by Gauss Jordon method
x  y  z  9 ; 2 x  3 y  4 z  13 ; 3x  4 y  5 z  40
Solution: Given x  y  z  9 ; 2 x  3 y  4 z  13 ; 3x  4 y  5 z  40
1 1 1 9  x
Let A   2 3 4 ; B  13  & X   y
     
 3 4 5   40   z 

1 1 1 : 9
 A : B   2 3 4 : 13 

 3 4 5 : 40 

DEPARTMENT OF MATHEMATICS, B.I.T

31
 Calculus & Linear Algebra (22MAT11)

1 1 1 : 9
R2  R2  2 R1 0 5 2 : 5
R3  R3  3R1  
0 1 2 : 13 

1 1 1 : 9
R2  R3 0 1 2 : 13 
 
0 5 2 : 5

1 1 9
1 :
R3  R3  5 R2 0 1 2 : 13 
 
0 0 12 : 60 

1 1 1 : 9
R
R3  3 0 1 2 : 13
12  
0 0 1 : 5 

1 1 0 : 4
R1  R1  R3 0 1 0 : 3
R2  R2  2 R3  
0 0 1 : 5 

1 0 0 : 1
R1  R1  R2 0 1 0 : 3
 
0 0 1 : 5 

1 
 X   3 X  1 3 5
T
 
Or
5

2) Solve by Gauss Jordon method

2 x  3 y  z  5 ; 4 x  4 y  3z  3 ; 2 x  3 y  2 z  2
Solution: Given 2 x  3 y  z  5 ; 4 x  4 y  3z  3 ; 2 x  3 y  2 z  2

DEPARTMENT OF MATHEMATICS, B.I.T

32
 Calculus & Linear Algebra (22MAT11)

 2 3 1 5  x
Let A   4 4 3 ; B  3 & X   y
     
 2 3 2   2  z 

2 3  1 : 5
 A : B   4 4  3 : 3

 2 3 2 : 2 

2 3 1 : 5
R2  R2  2 R1  0 2  1 : 7 
R3  R3  R1  
 0 6 3 : 3 

 2 3 1 : 5 
R 2   R 2  0 2 1 : 7 
R 3  R 3  3R 2  
 0 0 6 : 18 

 2 3 1 : 5
R  0 2 1 : 7 
R3   3 
 6   
 0 0 1 : 3 

 2 3 0 : 8
R1  R1  R3 0 2 0 : 4 
R2  R2  R3  
 0 0 1 : 3 

 2 3 0 : 8
R  0 1 0 : 2 
R2   2 
 2   
 0 0 1 : 3 

2 0 0 : 2
R1  R1  3R2 0 1 0 : 2
 
 0 0 1 : 3 

1 0 0 : 1
R  0 1 0 : 2 
R1   1 
 2  
0 0 1 : 3 

DEPARTMENT OF MATHEMATICS, B.I.T

33
 Calculus & Linear Algebra (22MAT11)

1 
 X  2  X  1 2 3
T
 
Or
 3 

3) Solve by Gauss Jordon method

2 x  5 y  7 z  52  0 ; 2 x  y  z  0 ; x  y  z  9  0
Solution: Given 2 x  5 y  7 z  52  0 ; 2 x  y  z  0 ; x  y  z  9  0
2 5 7  52   x
Let A   2 1 1 ; B0 & X   y
     
1 1 1   9   z 

2 5 7 : 52 
 A : B   2 1 1 : 0 

1 1 1 : 9 

1 1 1 : 9 
R2  R1  2 1 1 : 0 
 
 2 5 7 : 52 

2 5 7 : 52 
R2  R2  R1 2 1 1 : 0 
R3  R3  2 R1  
1 1 1 : 9 

 2 3 1 : 5 
R 2   R 2  0 2 1 : 7 
R 3  R 3  3R 2  
 0 0 6 : 18 

 2 3 1 : 5
R  0 2 1 : 7 
R3   3 
 6   
 0 0 1 : 3 

DEPARTMENT OF MATHEMATICS, B.I.T

34
 Calculus & Linear Algebra (22MAT11)

 2 3 0 : 8
R1  R1  R3 0 2 0 : 4 
R2  R2  R3  
 0 0 1 : 3 

 2 3 0 : 8
R  0 1 0 : 2 
R2   2 
 2   
 0 0 1 : 3 

2 0 0 : 2
R1  R1  3R2 0 1 0 : 2
 
 0 0 1 : 3 

1 0 0 : 1
R  0 1 0 : 2 
R1   1 
 2  
0 0 1 : 3 

1 
 X  2  X  1 2 3
T
 
Or
 3 

4) Solve by Gauss Jordon method

x  y  z  8 ;  x  y  2 z  4 ; 3x  5 y  7 z  14
Solution: Given x  y  z  8 ;  x  y  2 z  4 ; 3x  5 y  7 z  14
 1 1 1  8  x
Let A   1 1 2 ; B   4  & X   y
     
 3 5  7  14   z 

 1 1 1 : 8
 A : B    1 1 2 : 4 
 3 5 7 : 14 

1 1 1 9 
:
R2  R2  R1 0 0 3 : 4 
R3  R3  3R1  
0 2 10 : 10 

DEPARTMENT OF MATHEMATICS, B.I.T

35
 Calculus & Linear Algebra (22MAT11)

1 1 9
1 :
R  0
R2   3  1 5 : 5
 2   
0 0 3 : 4 
1 1 1 : 9 
R3  
R3  0 1 5 : 5 
3  
0 0 1 : 4 
3
1 1 0 : 20 
 3
R1  R1  R3 0 5 
1 0 :
R2  R2  5R3  3 
0 0 1 : 4 
 3 
1 0 0 : 5 
 
R1  R1  R2 0 1 0 : 5 
 3
0 0 1 : 4 
 3

5
 
 X  5 
T
Or  X  5 5 4 
 3  3 3
4 
 3
5) Solve by Gauss Jordon method
2x1  x2  3x3  1 ; 4x1  4x2  7 x3  1 ; 2 x1  5x2  9 x3  3
Solution: Given 2x1  x2  3x3  1 ; 4x1  4x2  7 x3  1 ; 2 x1  5x2  9 x3  3
2 1 3 1  x1 
Let A   4 4 7 ; B  1 & X   x2 
     
 2 5 9  3  x3 

DEPARTMENT OF MATHEMATICS, B.I.T

36
 Calculus & Linear Algebra (22MAT11)

 2 1 3 : 1
 A : B   4 4 7 : 1
 2 5 9 : 3

2 1 3 : 1
R2  R2  2 R1  0 2 1 : 1
R3  R3  R1  
 0 4 6 : 2 

2 1 3 : 1
R 3  R 3 2 R 2  0 2 1 : 1
 
 0 0 4 : 4 

2 1 3 : 1 
R   0 2 1 : 1
R3   3 
 4   
 0 0 1 : 1 

2 5 0 : 4
R1  R1  3R3 0 2 0 : 2 
R2  R2  R3  
 0 0 1 : 1 

2 5 0 : 4
R  0
R2   2  1 0 : 1
 2   
 0 0 1 : 1 

 2 0 0 : 1
R1  R1  5R2  0 1 0 : 1
 
 0 0 1 : 1 

1 0 0 : 1 
R   2
R1   1  0 1 0 : 1 
 2   
0 0 1 : 1


DEPARTMENT OF MATHEMATICS, B.I.T

37
 Calculus & Linear Algebra (22MAT11)

1 
 2 T
 X   1  Or  X   1 1 1
   2 
 1 

6) Solve by Gauss Jordon method

2 x  y  z  10 ; 3x  2 y  3z  18 ; x  4 y  9 z  16
Solution: Given 2 x  y  z  10 ; 3x  2 y  3z  18 ; x  4 y  9 z  16
2 1 1  10   x
Let A   3 2 3  ; B  18 & X   y
     
1 4 9  16   z 

 2 1 1 : 10 
 A : B   3 2 3 : 18 
1 4 9 : 16 

1 4 9 : 16 
R1  R3  3 2 3 : 18 
 
 2 1 1 : 10 

1 4 9 : 16 
R2  R2  3R1 0 10 24 : 30 
R3  R3  2 R1  
0 7 17 : 22 

R 2  R2
2  1 4 9 : 16 
0 5 12 : 15 
 
R3   R 3  0 7 17 : 22 

1 4 9 : 16 
7  
R3  R3    R2 0 5 12 : 15 
5  
0 0 15 : 5 5 

DEPARTMENT OF MATHEMATICS, B.I.T

38
 Calculus & Linear Algebra (22MAT11)

1 4 9 : 16 
R3  5R3 0 5 12 : 15 
 
0 0 1 : 5 

1 4 0 : 29 
R1  R1  9 R3 0 5 0 : 45 
R2  R2  12 R3  
0 0 1 : 5 

1 4 0 : 29 
R  0 1 0 : 9 
R2   2 
 5   
0 0 1 : 5 

1 0 0 : 7 
R1  R1  4 R2 0 1 0 : 9 
 
0 0 1 : 5 

 7
 X   9  X  7 9 5
T
 
Or
 5 

Numerical Methods
Numerical method provides various techniques to find approximate solution to
difficult problems using simple operations.
Iterative method
By iteration process we are finding the approximate solution of the given system of
equation. In this process in each step the result of previous step is used and the process is
carried out till we get the result to desired accuracy. The value obtained in each step is
always better than the previous.

Gauss Seidel Iterative Method:

Let us consider a system of 3 linear non-homogenous equations with 3 unknowns

DEPARTMENT OF MATHEMATICS, B.I.T

39
 Calculus & Linear Algebra (22MAT11)

a1 x  b1 y  c1 z  d1
a2 x  b2 y  c2 z  d 2
a3 x  b3 y  c3 z  d3
Let us consider x , y & z are the unknowns. Here the diagonal co-efficient
a1 , b2 & c3 should not be equal to zero and
a1  b1  c1 ; b2  a2  c2 ; c3  a3  b3 then this system of equations is said to
be diagonally dominant.

Working Rule
Step1: From the given system of equations
1
x  d1  b1 y  c1 z 
a1
1
y  d 2  a 2 x  c2 z 
b2
1
z  d3  a3 x  b3 y 
c3
Step2: Consider Initial values x0 , y0 , z0
Step3: First iteration is
1
x1   d1  b1 y0  c1 z0 
a1
1
y1   d 2  a2 x1  c2 z0 
b2
1
z1   d3  a3 x1  b3 y1 
c3
Step4: Second iteration is

DEPARTMENT OF MATHEMATICS, B.I.T

40
 Calculus & Linear Algebra (22MAT11)

1
x2   d1  b1 y1  c1 z1 
a1
1
y2   d 2  a2 x2  c2 z1 
b2
1
z2   d3  a3 x2  b3 y2 
c3
Step4: Third iteration is
1
x2   d1  b1 y1  c1 z1 
a1
1
y2   d 2  a2 x2  c2 z1 
b2
1
z2   d3  a3 x2  b3 y2 
c3
Step5: Fourth iteration is
1
x4   d1  b1 y3  c1 z3 
a1
1
y4   d 2  a2 x4  c2 z3 
b2
1
z4   d3  a3 x4  b3 y4 
c3
Step5: Fifth iteration is
1
x5   d1  b1 y4  c1 z4 
a1
1
y5   d 2  a2 x5  c2 z4 
b2
1
z5   d3  a3 x5  b3 y4 
c3

DEPARTMENT OF MATHEMATICS, B.I.T

41
 Calculus & Linear Algebra (22MAT11)

Problems:
1. Employing the Gauss Siedel method, solve the system of equations:
10 x  y  z  12 ; 2 x  10 y  z  13 ; 2 x  2 y  10 z  14
Solution:The given equations may be rewritten as

1
10 x  y  z  12  x  12  y  z
10
1
2 x  10 y  z  13  y  13  z  2 x
10
1
2 x  2 y  10 z  14  z  14  2 x  2 y
10
 0
Let the initial approximations be x  0 ; y  0  0 ; z  0  0

First iteration

 x  
1 1
10
 0 0

12  y    z    12   0    0   1.2
1
10
 y 1 
1
10
 
13  z  0  2 x 1  13   0   2 1.2   1.06
1
10
 z 1 
1
10
 
14  2 x1  2 y 1  14  2 1.2   2 1.06   0.948
1
10
Second iteration

 x  
2 1
10
 1 1

12  y    z    12  1.06    0.948   0.9992
1
10
 y  2 
1
10
 
13  z 1  2 x  2  13   0.948   2  0.9992   1.0054
1
10
 z  2 
1
10
 
14  2 x  2  2 y  2  14  2  0.9992   2 1.0054   0.9991
1
10
Third iteration

DEPARTMENT OF MATHEMATICS, B.I.T

42
 Calculus & Linear Algebra (22MAT11)

 x  
3 1
10
 2

12  y    z    12  1.0054    0.9991  0.9996
2 1
10
 y  3 
1
10
 
13  z  2   2 x  3  13   0.9991  2  0.9996   1.0002
1
10
 z  3 
1
10
 
14  2 x  3  2 y  3  14  2  0.9996   2 1.0002   1.000
1
10
2. Using the Gauss Siedel method, solve the equations:
2 x  3 y  20 z  25 ; 20 x  y  2 z  17 ; 3x  20 y  z  18
Solution:The given equations are not diagonally dominant, so rearranging it
20 x  y  2 z  17
3x  20 y  z  18
2 x  3 y  20 z  25

1
10 x  y  z  12  x  12  y  z
10
1
2 x  10 y  z  13  y  13  z  2 x
10
1
2 x  2 y  10 z  14  z  14  2 x  2 y
10
 0
Let the initial approximations be x  0 ; y  0  0 ; z  0  0

First iteration

 x  
1 1
10
 0 0

12  y    z    12   0    0   1.2
1
10
 y 1 
1
10
 
13  z  0  2 x 1  13   0   2 1.2   1.06
1
10
 z 1 
1
10
 
14  2 x1  2 y 1  14  2 1.2   2 1.06   0.948
1
10
Second iteration

DEPARTMENT OF MATHEMATICS, B.I.T

43
 Calculus & Linear Algebra (22MAT11)

 x  
2 1
10
 1

12  y    z    12  1.06    0.948   0.9992
1 1
10
 y  2 
1
10
 
13  z 1  2 x  2  13   0.948   2  0.9992   1.0054
1
10
 z  2 
1
10
 
14  2 x  2  2 y  2  14  2  0.9992   2 1.0054   0.9991
1
10
Third iteration

 x  
3 1
10
 2

12  y    z    12  1.0054    0.9991  0.9996
2 1
10
 y  3 
1
10
 
13  z  2   2 x  3  13   0.9991  2  0.9996   1.0002
1
10
 z  3 
1
10
 
14  2 x  3  2 y  3  14  2  0.9996   2 1.0002   1.000
1
10
3. Apply Gauss Siedel method, solve the system of equations:
5 x  2 y  z  12 ; x  4 y  2 z  15 ; x  2 y  5 z  20
Carryout 4 iterations, taking the initial approximation to the solution as
1, 0,3
Solution:The given equations may be rewritten as

1
5 x  2 y  z  12  x  12  2 y  z
5
1
x  4 y  2 z  15  y  15  2 z  x
4
1
x  2 y  5 z  20  z  20  x  2 y
5
 0
Given the initial approximations to the solution is x  1 ; y  0  0 ; z  0  3

First iteration

DEPARTMENT OF MATHEMATICS, B.I.T

44
 Calculus & Linear Algebra (22MAT11)

 x  
1 1
5
 0

12  2 y    z    12  2  0    3  1.8
0 1
5
1
4
 
 y 1  15  2 z  0  x 1  15  2  3  1.8   1.8
1
4
1

5
 1 1

 z    20  x   2 y    20  1.8   2 1.8   2.92
1 1
5
Second approximation

 x  
2 1
5
 1

12  2 y    z    12  2 1.8    2.92   1.096
1 1
5
1
4
 
 y  2  15  2 z 1  x  2  15  2  2.92   1.096   2.016
1
4
2

5
 2

 z    20  x   2 y    20  1.096   2  2.016   2.9744
1 2 1
5
Third approximation

 x  
3 1
5
 2

12  2 y    z    12  2  2.016    2.9744   0.9987
2 1
5
1
4
 
 y  3  15  2 z  2  x  3  15  2  2.9744    0.9987   2.0131
1
4
3

5
 3

 z    20  x   2 y    20   0.9987   2  2.0131  2.9950
1 3 1
5
Fourth approximation

 x  
4 1
5
 3

12  2 y    z    12  2  2.0131   2.9950   0.9958
3 1
5
1
4
 
 y  4  15  2 z  3  x  4  15  2  2.9950    0.9958   2.0036
1
4
4

5
 4

 z    20  x   2 y    20   0.9958   2  2.0036   2.9994
1 4 1
5

the solution of the given system is x  0.9958 ; y  2.0036 ; z  2.9994

DEPARTMENT OF MATHEMATICS, B.I.T

45
 Calculus & Linear Algebra (22MAT11)

4. Apply Gauss Siedel method, solve the system of equations:

x  y  54 z  110 ; 27 x  6 y  z  85 ; 6 x  15 y  2 z  72

Eigen values and eigen vectors:

For a given square matrix ' A ' if  a scalar  and a nonzero column matrix X such
that A X   X .Then the scalar  is called as eigen value of the matrix ' A ' and X is
called the eigen vector of ' A ' corresponding to the eigen value  .
AX   X
AX  I X
 AX I X  0

  A  I  X  0
This represents a system of ' n ' homogenous equation in variables ' n ' variables
x1 , x2 , x3 .............xn . This system of equation has non trivial solution if the co-efficient
matrix  A   I  is singular.

i.e., A   I  0

a 11  a 12 a13 a 1n
a21 a2 2   a 2 3 a2 4
 0

am1 a m2 am 3 am n  

Expansion of the determinant gives a nth degree of polynomial which is known as

characteristic polynomial of ' A ' and A   I  0 is known as characteristic equation of
' A' .
Note:
rd
1. The characteristic equation of a 3 order square matrix can e obtained expanding
A   I  0 i.e.,   3    d   2    m d    A  0
Where   d  Sum of diagonal elements

DEPARTMENT OF MATHEMATICS, B.I.T

46
 Calculus & Linear Algebra (22MAT11)

 md  Sum of minor of diagonal elements

A  Determinant of ' A '
2. Give 3 vectors, if one vector is a multiple of the other than the vectors are linearly
dependent.
Eg: X 1  1 0 3 ; X 2   2 0 6  ; X 3   1 0 3
T T T

 X1 , X 2 , X 3 are linearly dependent.

3. If one vector is not the multiple of the other then the vectors are linearly
independent.

Property:
1. Sum of the eigen values of a square matrix is equal to the sum of the principle
diagonal elements.
2. Product of the eigen values of a square matrix is equal to the determinant of the
matrix.

Problems:
Find the eigen value and eigen vector of the following matrices

1. 5 4
1 2 
 
5 4 
Solution: Let A 
1 2 
Characteristic equation is given by A   I  0

5 4
  0
1 2
  5    2     4  0
  10  5   2    2   4  0
  2 7   6  0
    6    1   0
  1 , 6

DEPARTMENT OF MATHEMATICS, B.I.T

47
 Calculus & Linear Algebra (22MAT11)

The roots of the characteristic equation are  1  1 &  2  6.

Consider  A   I  X  0

5   4   x  0
    0
 1 2     y  
5    x  4 y  0
 1
x  2    y  0

Case 1: If  1  1 in equation 1 then

4 x 4 y  0
1 
x y  0
 x y  0
 x  y
x y
 
1 1
 X   1 1 
T

Case 2: If 26 in equation 1 then

x  4 y  0
1 
x 4 y  0
 x  4y  0
 x  4y
x y
 
4 1
 X  4 1
T

DEPARTMENT OF MATHEMATICS, B.I.T

48
 Calculus & Linear Algebra (22MAT11)

 8 6 2 
2.  6 7 4 
 
 2 4 3 
 8 6 2 
A   6 7 4 
Solution: Let
 
 2 4 3
Characteristic equation is given by A   I  0

3   d 2   m d   A  0
 7 4 8 2 8 6 
  3   8  7  3  2       0  0
  4 3 2 3  6 7 
  3  18   2    21  16    24  4    56  36     0
  3  18 2   5  20  20    0
  3  18 2  45  0
    2  18  45   0
    2  3  15  45   0
      3   15    3   0
      3   15    0
   0,3,15
The roots of the characteristic equation are  1  0 ,  2  3 &  3  15 .
Consider  A   I  X  0

DEPARTMENT OF MATHEMATICS, B.I.T

49
 Calculus & Linear Algebra (22MAT11)

 8 6 2   x 0
  6 7   4   y   0
    
 2 4 3     z  0
8    x  6 y  2 z  0
 6 x   7    y  4 z  0 1
2x  4 y  3    z  0

Case 1: If  1  0 in equation 1 then

8 x  6 y  2z  0 i 
1  6 x  7 y  4 z  0  ii 
2 x  4 y  3z  0  iii 

x y z
From  i  &  ii    
6 2 8 2 8 6
7 4 6 4 6 7
x y z
  
 24  14   32  12   56  36 
x y z
   
10 20 20
x y z
  
1 2 2
Case 2: If  2  3 in equation 1 then
5 x  6 y  2z  0 i 
1  6 x  4 y  4 z  0  ii 
2x  4 y  0z  0  iii 

DEPARTMENT OF MATHEMATICS, B.I.T

50
 Calculus & Linear Algebra (22MAT11)

x y z
From  ii  &  iii    
4 4 6  4 6 4
4 0 2 0 2 4
x y z
   
 0  16   0  8  24  8
x y z
   
16 8 16
x y z
  
2 1 2

Case 3: If  3  15 in equation 1 then

7 x  6 y  2 z  0 i 
1  6 x  8 y  4 z  0  ii 
2 x  4 y  12 z  0  iii 

x y z
From  i  &  ii    
6 2 7 2  7 6
8  4 6 4 6 8
x y z
  
 24  16   28  12   56  36 
x y z
   
40 40 20
x y z
  
2 2 1

DEPARTMENT OF MATHEMATICS, B.I.T

51
 Calculus & Linear Algebra (22MAT11)

 6 2 2
3.  2 3 1
 
 2 1 3
 6 2 2
A   2 3 1
Solution: Let
 
 2 1 3 
Characteristic equation is given by A   I  0

3   d 2   m d   A  0
 3 1 6 2 6 2 
  3   6  3  3  2        32  0
  1 3 2 3  2 3 
  3  12   2    9  1  18  4   18  4     32  0
  3  12 2   8  14  14    32  0
  3  12 2  36  32  0
 2 , 2 , 8
The roots of the characteristic equation are  1   2  2 &  3  8.
Consider  A   I  X  0

6   2 2   x 0
  2 3   1   y   0
    
 2 1 3     z  0
6    x  2 y  2z  0
 2 x   3    y  z  0 1
2x  y  3    z  0

Case 1: If  1   2 2 in equation 1 then

DEPARTMENT OF MATHEMATICS, B.I.T

52
 Calculus & Linear Algebra (22MAT11)

4 x  2 y  2z  0 i 
1  2 x  y  z  0  ii 
2x  y  z  0  iii 

From  i  , ii  & iii  the equations are linearly dependent.  2 x  y  z  0 gives an infinite
4 2 2 : 0

solution. Here augumented matrix is  A : B   2 1  1 : 0
 
 2 1 1 : 0 

Where  1 ; n  3   n let us choose n    3  1  2 arbitrary constants i.e., let

k2  k1
z  k1 & y  k2 then from 2 x  y  z  0  x 
2

k  k
T

 For   2, X   2 1 k2 k1 
 2 

Case 2: If  2  8 in equation 1 then

2 x  2 y  2 z  0 i 
1  2 x  5 y  z  0  ii 
2 x  y  5z  0  iii 

DEPARTMENT OF MATHEMATICS, B.I.T

53
 Calculus & Linear Algebra (22MAT11)

x y z
From  i  &  ii    
2 2 2 2  2 2
5 1  2 1 2 5
x y z
  
 2  10   2  4  10  4 
x y z
   
12 6 6
x y z
  
2 1 1
 For   2 , X   2 1 1
T

 1 1 2
4.  0 2 2
 
  1 1 3
 1 1 2
A   0 2 2
Solution: Let
 
 1 1 3 
Characteristic equation is given by A   I  0

3   d 2   m d   A  0
 2 2 1 2 1 1
  3  1  2  3  2       6  0
 1 3  1 3 0 1 
  3   6   2    6  2    3  2   1  0     6  0
  3  6 2   4  5  1   6  0
  3  6 2  10  6  0
  1 , 2 , 3
The roots of the characteristic equation are  1  1 ,  2  2 &  3  3.

DEPARTMENT OF MATHEMATICS, B.I.T

54
 Calculus & Linear Algebra (22MAT11)

Consider  A   I  X  0

 1  1 2   x 0 
 0 2 2   y  0
    
  1 1 3     z  0
1    x  y  2 z  0
 0x   2    y  2 z  0 1
 x  y  3    z  0

Case 1: If  1  1 in equation 1 then

0x  y  2z  0 i 
1  0x  y  2 z  0 ii 
x  y  2z  0 iii 
Since  i  ,  ii  are linearly dependent, let us choose  ii  ,  iii  for cross multiplication
x y z
From  ii  &  iii    
1 2 0 2 0 1
1 2 1 2 1 2
x y z
  
 2  2   0  2   0  1
x y z
   
0 2 1
x y z
  
0 2 1
 For   1 , X   0 2 1
T

Case 2: If  2  2 in equation 1 then

x  y  2z  0 i 
1  0x  0 y  2 z  0  ii 
x  y  2z  0  iii 

DEPARTMENT OF MATHEMATICS, B.I.T

55
 Calculus & Linear Algebra (22MAT11)

Since  i  ,  iii  are linearly dependent, let us choose i  , ii  for cross multiplication

x y z
From  i  &  iii    
1 2 1 2 1 1
0 2 0 2 0 0
x y z
   
 2  0  2  0  0  0
x y z
   
2 2 0
x y z
  
1 1 0
 For   2 , X  1 1 0
T

Case 3: If  3  3 in equation 1 then

2 x  y  2 z  0 i 
1  0x  y  2 z  0  ii 
x  y  0z  0 iii 

x y z
From  i  &  ii    
1 2 2 2 2 1
1 2 0 2 0 1
x y z
  
 2  2   4  0   2  0 
x y z
   
4 4 2
x y z
  
2 2 1
 For   3 , X   2 2 1
T

DEPARTMENT OF MATHEMATICS, B.I.T

56
 Calculus & Linear Algebra (22MAT11)

 2 2  3
5.  2 1 6 
 
 1 2 0 
 2 2  3
A   2 1 6 
Solution: Let
 
 1 2 0 
Characteristic equation is given by A   I  0

3   d 2   m d   A  0
 1 6 2 3 2 2 
  3   2  1  0   2        45  0
 2 0  1 0 2 1 
    1     0  12    0  3   2  4     45  0
3 2

  3   2   12  3  6    45  0
  3   2  21  45  0
  5 , 3 , 3
The roots of the characteristic equation are  2   3   3 &  1  5.
Consider  A   I  X  0

 2   2  3   x 0
  2 1   6   y    0 
     
 1 2 0     z  0
 2    x  2 y  3 z  0
 2 x  1    y  6 z  0 1
x  2 y  0    z  0

Case 1: If  1  5 in equation 1 then

DEPARTMENT OF MATHEMATICS, B.I.T

57
 Calculus & Linear Algebra (22MAT11)

7 x  2 y  3z  0 i 
1  2x  4 y  6z  0 ii 
 x  2 y  5z  0  iii 

x y z
From  i  &  ii    
2 3 7  3 7 2
4 6 2 6 2 4
x y z
  
 12  12   42  6   28  4 
x y z
   
24 48 24
x y z
  
1 2 1
 For   5 , X  1 2 1
T

Case 2: If  2   3  3 in equation 1 then

x  2 y  3z  0 i 
1  2x  4 y  6z  0  ii 
 x  2 y  3z  0  iii 

From i  , ii  & iii  the equations are linearly dependent.  x  2 y  3z  0 gives an infinite
1 2  3 : 0

solution. Here augumented matrix is  A : B   2 4  6 : 0
 
 1 2 3 : 0 

Where  1 ; n  3   n let us choose n    3  1  2 arbitrary constants i.e., let

z  k1 & y  k2 then from x  2 y  3z  0  x  3k1  2k2

 For   3, X  3k1  2k2 k1 

T
k2

DEPARTMENT OF MATHEMATICS, B.I.T

58
 Calculus & Linear Algebra (22MAT11)

Rayleigh’s Power Method:

This is an iterative method to determine the numerically largest eigen value (dominant eigen value) &
the corresponding eigen vector of a square matrix.
Working rule:
If ' A' is a given square matrix
1  0  0  1
      
1. Consider the initial eigen vector X 0  0 or X 0  1 or X 0  0 or X 0  1
      
0  0  1  1
2. Take out the numerically largest value as the common factor to obtain or A X 0  1 X1 . This
method is called as normalization.
3. Find A X 1 and again put it in the formA X1  2 X 2 by normalization.
4. Continue the same procedure till two consecutive iterative values of  are same up to a desired
accuracy.

DEPARTMENT OF MATHEMATICS, B.I.T

59
 Calculus & Linear Algebra (22MAT11)

Problems:
1. Find the largest eigen value and corresponding eigen vector by power method
 25 1 2 
 1 3 0
 
 2 0 4
 25 1 2  1 
A   1 3 0 X 0  0 
Solution: Let
&
   
 2 0 4 0
 25 1 2 1   25  1 
A X0   1 3 0 0   1   25 0.04   1 X 1
       
 2 0 4  0   2   0.08
 25 1 2 1   25.2   1 
A X1   1 3 
0 0.04    1.12   25.2 0.0444   2 X 2
      
 2 0 4   0.08  1.68  0.0666 
 25 1 2   1   25.1776   1 
A X2   1 3 0  0.0444    1.1332   25.1776 0.0450   3 X 3
      
 2 0 4  0.0666   1.7336  0.0689 
 25 1 2   1   25.1828  1 
A X3   1 3 0  0.0450    1.135   25.1828  0.0451  4 X 4
      
 2 0 4  0.0689   1.7244  0.0685
 25 1 2   1   25.1828  1 
A X4   1 3 0  0.0450    1.135   25.1821  0.0451  5 X 5
      
 2 0 4  0.0689   1.7244 
    0.0685
The largest eigen value is   25.1821 and the corresponding eigen vector is

DEPARTMENT OF MATHEMATICS, B.I.T

60
 Calculus & Linear Algebra (22MAT11)

 1 
 0.0451 or 1 0.0451 0.0685 T .
   
0.0685

2. Find the largest eigen value and corresponding eigen vector by taking the initial
4 1  1
approximation to the eigen vector as1 0.8 0.8 T .Perform 5 iterations 
2 3  1
 
.
  2 1 5 

 4 1  1  1 
A   2 3  1 X 0   0.8 
Solution: Given
&
   
  2 1 5   0.8
4 1  1  1   5.6   0.8235 
A X0   2 3  
 1 0.8  6.8     6.8  1   1 X 1
      
 2 1 5   0.8  5.2   0.7647 
4 1  1  0.8235   5.0587   0.9348 
A X1   2 3  1   1    5.4117   5.4117  1   2 X 2
       
 2 1 5   0.7647   4.4705  0.8261
4 1  1  0.9348   5.5653   0.9771 
A X2   2 3  1   1    5.6957   5.6957  1   3 X 3
      
 2 1 5   0.8261  5.0001  0.8838
4 1  1  0.9771 
A X3   2 3  1   1         4 X 4
  
 2 1 5   0.8838
4 1  1
A X4   2 3  1         5 X 5
 
 2 1 5 
The largest eigen value is   25.1821 and the corresponding eigen vector is

DEPARTMENT OF MATHEMATICS, B.I.T

61
 Calculus & Linear Algebra (22MAT11)

 1 
 0.0451 or 1 0.0451 0.0685 T .
   
0.0685

3. Find the largest eigen value and corresponding eigen vector corrected to 3 decimal
 2 1 0 
places by taking the initial eigen vector as1 1 1 T where
A   1 2 1
 
 0 1 2 
 2 1 0  1
A  1 2 1 X 0  1
Solution: Given
&
  
 0 1 2  1

DEPARTMENT OF MATHEMATICS, B.I.T

62
 Calculus & Linear Algebra (22MAT11)

 2 1 0  1 1  1 
A X 0   1 2 1 1  0   1 0   1 X 1
     
 0 1 2  1 1  1 
 2 1 0  1   2  1
A X 1   1 2 1 0    2   2  1  2 X 2
       
 0 1 2  1   2   1 
 2 1 0  1  3   0.75 
A X 2   1 2 1 1  4  4  1   3 X 3
    
     
 0 1 2  1  3   0.75
 2 1 0   0.75   2.5   0.7143 
A X 3   1 2 1  1    3.5  3.5  1   4 X 4
      
 0 1 2   0.75  2.5   0.7143
 2 1 0   0.7143   2.4286   0.7085 
A X 4   1 2 1  1    3.428   3.428  1   5 X 5
      
 0 1 2   0.7143  2.4286   0.7085
 2 1 0   0.7085   2.417   0.7073 
A X 5   1 2 1  1    3.417   3.417  1   6 X 6
      
 0 1 2   0.7085  2.417   0.7073
 2 1 0   0.7073   2.4146   0.7071 
A X 6   1 2 1   1   3.4146  3.4146  1   7 X 7
 
      
 0 1 2   0.7073  2.4146   0.7071

 2 1 0   0.7071   2.4142   0.7071 

A X 7   1 2 1  1    3.4142  3.4142  1   8 X 8
      
 0 1 2   0.7071  2.4142   0.7071

The largest eigen value is   3.4142 and the corresponding eigen vector is

DEPARTMENT OF MATHEMATICS, B.I.T

63
 Calculus & Linear Algebra (22MAT11)

 0.7071 
 1  or 0.7071 1 0.7071 T .
   
 0.7071

Reduction to diagonal matrix:

Similarity of matrices:
' A' and ' B ' of the same order are said to be similar if there exist a non singular ' P ' such that
2 matrices
1
B  P A P .Here ' B ' is similar to ' A' .
Diagonlisation:
A square matrix ' A' is said to be diagonalizable if it is similar to a diagonal matrix.

Property:
' A' of order ' n ' has n  linearly independent eigen vectors, then a matrix ' P ' can be
If a square matrix
1
found such that P A P  D is a diagonal matrix.
Where D = diagonal matrix also known as spectral matrix
P = matrix which transform ' A' to diagonal form is known as modal matrix

Note:
1. The transformation of a square matrix ' A' to P 1 A P is known as similarity transformation.
2. The diagonal matrix has the eigen values of ' A ' as its diagonal elements.
3. The matrix ' P ' which diagonalise the square matrix, constitute the eigen vector.

Computation of powers of a square matrix:

Let D be a square matrix. Then a non singular matrix P can be found such that

DEPARTMENT OF MATHEMATICS, B.I.T

64
 Calculus & Linear Algebra (22MAT11)

D  P 1 A P
 D  P 1 A P
 D 2   P 1 A P  P 1 A P 
 D 2  P 1 A  PP 1  A P
 D 2  P 1 A  I  A P
 D 2  P 1 A 2 P
Pre multiplying by P and post multiplying by P 1
 PD 2 P 1  PP 1 A 2 PP 1
 PD 2 P 1  I A 2 I
 PD 2 P 1  A 2
 PD 3 P 1  A 3
1
In general PD P
n
 An

Note:
a b  1 1  d b 
If A    
A  c a 
then its inverse is given by A .
c d 

Problems:

1. Reduce the matrix  1 3  to the diagonal form and hence find A 4

 2 4
 
 1 3
Solution: Let A 
 2 4
Characteristic equation is given by A   I  0

DEPARTMENT OF MATHEMATICS, B.I.T

65
 Calculus & Linear Algebra (22MAT11)

1   3
  0
2 4
   1    4     6  0
   4   4    2  6  0
  2 3   2  0
    2    1   0
  1 , 2
The roots of the characteristic equation are 11 &  2  2 .

Consider  A   I  X  0

 1   3   x  0
    0
 2 4     y   
 1    x  4 y  0
 1
x  4    y  0

Case 1: If  1  1 in equation 1 then

2 x  3 y  0
1 
2 x  3 y  0
  2x  3y  0
 2x  3y
x y
 
3 2
 X  3 2
T

Case 2: If  2  2 in equation 1 then

DEPARTMENT OF MATHEMATICS, B.I.T

66
 Calculus & Linear Algebra (22MAT11)

3x  3 y  0
1 
2 x  2 y  0
 x y  0
 x  y
x y
 
1 1
 X  1 1
T

The matrix which reduces given matrix to diagonal form is modal matrix

 3 1
Modal matrix P   
 2 1

1  1 1
P 1  
2  2 3 

1 1  1 1  1 3   3 1
Diagonal matrix D  P AP 
2  2 3   2 4   2 1

1 0 
D  P 1 AP   
0 2 

 3 1   1  3  1  1
5 5
1 1
A  PD P     
5 5

 2 1  2 
5
 4   2  2
5
3 

 61 93
A5   
 62 94 

2. Diagonalize the matrix  19 7

 42 16
 
Solution: Let A   19 7 
 42 16 
 

DEPARTMENT OF MATHEMATICS, B.I.T

67
 Calculus & Linear Algebra (22MAT11)

Characteristic equation is given by A   I  0

19   7
  0
42 16  
   19    16     294  0
  2  3   304  294  0
  2  3   10  0
    2    5  0
   2 , 5
The roots of the characteristic equation are 1 2 &  2  5 .

Consider  A   I  X  0

 19   7   x 0
    0
 42 16     y   
 19    x  7 y  0
 1
 42 x  16    y  0

Case 1: If  1  2 in equation 1 then

21 x  7 y  0
1 
42 x 14 y  0
  21x  7 y  0
 21x  7 y
 3x  y
x y
 
1 3
 X  1 3
T

Case 2: If  2   5 in equation 1 then

DEPARTMENT OF MATHEMATICS, B.I.T

68
 Calculus & Linear Algebra (22MAT11)

14 x  7 y  0
1 
42 x  21 y  0
  14 x  7 y  0
 14 x  7y
 2x  y
x y
 
1 2
 X  1 2
T

The matrix which reduces given matrix to diagonal form is modal matrix

1 1 
Modal matrix P   
3 2 

 2 1  2 1 
P 1   1     3 1
 3 1   

1  2 1   19 7 1 1 
Diagonal matrix D  P AP     42
 3 1  16  3 2 
 

2 0 
D  P 1 AP   
 0 5

DEPARTMENT OF MATHEMATICS, B.I.T

69