1.

SOLUTION OF EQUATIONS AND EIGEN VALUE PROBLEMS

1.1 Method of False Position (or) Regula Falsi Method

Consider the equation f(x) = 0 and let f(a) and f(b) of opposite signs. Also, let a<b.

The curve y = f(x) will meet the x-axis at some point between A[a, f(a)] and B[b, f(b)]

The equation of the chord joining the two points A[a, f(a)] and B[b, f(b)] is

y  f (a ) f (a )  f (b)

xa a b

The x- coordinate of the point of intersection of this chord with the x-axis gives an
approximate value for the root of f(x) =0 setting y=0 in the chord equation, we get

 f (a ) f (a )  f (b)

xa a b

x[f(a) - f(b)] – af(a) + af(b) = -af(a) + bf(b)

x[f(a) – f(b)] = bf(a) – af(b)

af (b)  bf (a )
x1 
f (b)  f (a )

This value of x1 gives an approximate value of the root of f(x) = 0. (a<x1<b)

Now, f(x1) and f(b) are of opposite signs.
If f(x1), f(a)<0, then x2 lies between x1 and a.

af ( x1 )  x1 f (a )
x2 
Hence, f ( x1 )  f (a )

In the same way, we get x3, x4………..

This sequence x1, x2, x3… Will converge to the required root. In practice, we get xi and xi+1 such
x x 
that i i 1 , the required accuracy,

NOTE:

(i) The Convergence of the root in this method is a slower than the Newton-
Raphson method.

(ii) This method is also known as ‘Method of chords’.

EXAMPLE: 1

Find a real root of x  9 x  1  0 that lies between 2 and 3 by

3

the method of false position, correct to 3 decimal places.

SOLUTION:

Let f x   x 3  9 x  1
f 2   8  18  1  9  0
f 2.5  5.875  0

and f 3  27  27  1  0

Hence the root lies between a = 2.5 and b = 3.

The approximations are given by

a f b   b f a 
x
f b   f a 

First approximation:

Taking a = 2.5 and b = 3

2.51  3 5.875
x1   2.9273
1  5.875
and f(2.9273) = - 0.2664<0

Second approximation:

Taking a = 2.9273 and b = 3

2.92731  3 0.2664
x2   2.9423
1  0.2664
f (2.9423) = - 0.0088 < 0

Third approximation:

Taking a = 2.9423 and b = 3

2.94231  3 0.2664
x3   2.9425
1  0.0088
f (2.9425) = -0.0054 < 0

Fourth approximation:

Taking a = 2.9425 and b = 3

 2.94251  3 0.0054 
x4   2.943
1  0.0054
f (2.943) = 0.003 > 0

Since f (2.943) = 0.003 to nearly zero, the required root is 2.943

EXERCISE 1.1

1. Find the real root of x3 - 2x -5 = 0 correct to five decimals by the method of

false position.

2. Determine the real root of xex = 3 correct to four decimal places by Regula
Falsi method.

3. Find the positive real root of x = 2sinx, correct to four decimals by the method
of false position.

1.2 Newton’s Method (or) Newton – Raphson Method (or) Method of Tangent

This method starts with an initial approximation to the root of an equation, a better and
closer approximation to the root can be found by using an iterative process.

1.2.1 Derivation of Newton – Raphson Formula

Let α be a root to f(x) = 0 and x0 be an approximation to α. If h = α-x0, then by Taylor’s

series.
h2
f ( x0  h)  f ( x0 )  hf ( x0 )  f ( x0 )  .....
2 …… ------------ (i)

Since α = x0 + h is the root, f(x0 + h) = 0.

Neglecting the higher powers of h in (i), we find

f ( x0 )
h
f ( x0 )

Since some terms were omitted in (i), x1 = x0 + h is exactly α, but it is better

approximation than x0.

f ( x0 )
 x1  x0  h  x0 
f ( x0 )
Repeating the process, we get the iteration formula

f ( xn )
xn 1  xn 
f ( xn ) , n = 0,1,2……. -----------(ii)
 This is called Newton-Raphson formula.

EXAMPLE: 2

x log10 x  1.2,
Find by Newton’s method, the real root of correct to 4
decimal places.

Solution:

f x   x log10 x  1.2
Let

f
1
x   log10 x  x d log10 x
Then dx

d
 log10 x  x log e x log10 e
dx
1
 log10 x  x  log10 e  log10 x  log10 e
x
f x   log10 x  0.4343
1

f 2   0.5979  0
f 3  0.2314  0

The root lies between 2 and 3.

Choose the initial approximation x0 = 2.5

First approximation:
f x 0    0.2051 
x1  x0   2. 5     2.7465
f 1 x 0   0.8322 

Second approximation:
f x1   0.0051
x 2  x1   2.7465   2.7406
f 1 x1  0.8731

Third approximation:
 f x 2    0.00004 
x3  x 2   2.7406     2.7406
f 1 x 2   0.8721 

The required root is 2.7406

EXERCISE 1.2

1. Find by Newton's method, the real positive root of x = cosx, correct to 3 decimal
places.

2. Using Newton's method, find the root between 0 and 1 of x3 - 6x + 4 = 0, correct to

five decimal places.

3. Solve the equation e-x = sinx for the root lying between 0 and 1 by Newton Raphson
method.

1.3 Gauss-Elimination method

This is an elimination method and it reduces the given system of equation to an

equivalent upper triangular system which can be solved by back substitution.

Consider the system of equations

a11 x1 + a12 x2 + a13 x3 = b1
a21 x1 + a22 x2 + a23 x3 = b2
a31 x1 + a32 x2 + a33 x3 = b3

Gauss-algorithm is explained below:

Step: 1

Elimination of x1 from the second and third equations. If a11 ≠ 0, the first
equation is used to eliminate x1 from the second and third equation. After
elimination, the reduced system is
a11 x1 + a12 x2 + a13 x3 = b1
a’22 x2 + a’23 x3 = b’2
a’32 x2 + a’33 x3 = b’3

Step: 2

Elimination of x2 from the thied equation. If a’22 ≠ 0, we eliminate x2 from the

third equation and the reduced upper triangular system is
 a11 x1 + a12 x2 + a13 x3 = b1
a’22 x2 + a’23 x3 = b’2
a’’33 x3 = b’’3

Step: 3

From third equation x3 is known. Using x3 in the second equation, x2 is obtained.

Using both x2 and x3 in the first equation, the value of x3 obtained.

Thus in Gauss- elimination method, we start with the augmented matrix (A/B) of
the given system and transform it to (U/K) by elementary row operations. Finally
the solution is obtained by back substitution process.

EXAMPLE: 3

Solve the system of liner systems by Gauss elimination method

5x – y – 2z = 142; x – 3y – z = –30; 2x – y – 3z = 5.

Solution:

The augmented matrix of the given system is

 1 - 3 - 1 - 30 
 
 2 -1 - 3 5 
 5 - 1 - 2 142 
[A, B] =  

1 - 3 -1 - 30 
 
0 5 -1 65  R2  R2  2 R1
 0 14 3 292  R3  R3  5 R1
~
1 - 3 -1 - 30 
 
0 5 -1 65 
14
0 0 29/5 110  R3  R3  5 R2
~
The equivalent system is
x- 3y - z = -30
5y - z = 65
 29
z  110
5
5
z  110   18.966
Þ 29
65  18.966
y  16.793
5
x = -30 + 3 (16.793) + (18.966) = 39.345
The solution is
x = 39.345, y = 16.793, z = 18.966.

EXERCISE 1.3

SOLVE THE FOLLOWING EQUATIONS BY GAUSS ELIMINATION METHOD:

1. 2x + y + 4z = 12; 8x - 3y + 2z = 20; 4x + 11y - z =33
2. x + y +2z = 4; 3x + y -3z = -4; 2x -3y -5z = -5
3. x + y+ z =9; 2x -3y +4z =13; 3x +4y+ 5z =40.

1.4 Gauss-Jordan method

This method is a modification of Gauss-elimination method. Here the elimination

of unknowns is performed not only in the equations below but also in the equations
above. The co-efficient matrix A of the system AX = B is reduced into a diagonal
or a unit matrix and the solution is obtained directly without back substitution
process.

EXAMPLE:4

Solve by Gauss–Jordan method:

3x + 4y + 5z = 18; 2x – y + 8z = 13; 5x – 2y + 7z = 20.

Solution:

The augmented matrix of the given system is

3 4 5 18 
 
 2 -1 8 13 
5 - 2 7 20 
[A, B] = 
 1 5 - 3 5
 
 2 -1 8 13 
5 - 2 7 20  R1  R1  R2
~ 

1 5 - 3 5
 
 0 - 11 14 3  R12  R2  2 R1
 0 - 27 22 - 5  R3  R3  5 R1
~
1 5 3 5 
 
0 1  14 / 11  3 / 11  1
 0  27 22  5  R2   11 R2
~
 1 0 37/11 70/11 
 
 0 1  14 / 11  3 / 11  R1  R2  5 R2
 0 0  136/11  136 / 11  R3  R3  27 R2
~ 
 1 0 37/11 70/11 
 
 0 1  14 / 11  3 / 11  11
0 0 1 1  R3   136 R3
~
37
1 0 0 3  R1  R1  11 R3
 
0 1 0 1 14
0 0 1 1  R2  R2  11 R3
~

The solution is x = 3, y = 1, z = 1.

EXERCISE 1.4

SOLVE THE FOLLOWING EQUATIONS BY GAUSS JORDAN METHOD:

1. 2x -7y +4z = 9; x +9y-6z = 1; -3x +8y+5z = 6

2. x+2y+z =8; 2x+3y+4z = 20; 4x+y+2z =12;
3. 3x+4y-7z =23; 7x-y+2z =-14; x+10y-2z =33;

1.5 Gauss-Jocobi’s iteration method

Let the given system of simultaneous equations be

a1x + b1y + c1z = d1

a2x + b2y + c2z = d2 …………………………………….. (1)
a3x + b3y + c3z = d3

This system of equations can also be written as

1
x d1  b1 y  c1 z 
a1
1
y d 2 a2 x  c2 z 
b2
1
z d3  a3 x  b3 y 
c1 ………………………….(2)

Let the first approximation be x0, y0 and z0 substituting x0, y0 and z0 in (2), we get
1
x1  d1  b1 y0  c1 z0 
a1
1
y 1 d 2 a2 x0  c2 z0 
b2
1
z 1 d3  a3 x0  b3 y0 
c1

Substituting he values of x1, y1, z1 in (2) we get the 2nd approximations x2, y2 and
z2 as given below
1
x2  d1  b1 y1  c1 z1 
a1
1
y 2 d 2 a2 x1  c2 z1 
b2
1
z 2 d3  a3 x1  b3 y1 
c1

Substituting the values of x2, y2 and z2 in (2) we get the third approximations x3,
y3 and z3. This process may be repeated till the difference between two consecutive
approximations is negligible.

EXAMPLE:5
Solve the following system by Gauss-Jacobi method:
10x – 5y -2z = 3; x + 6y +10z = -3 ; 4x – 10y +3z = -3;

Solution:

The given equations are not diagonally dominant, Therefore, we rewrite the
equations as follow:

10x – 5y -2z = 3
4x – 10y +3z = -3
x + 6y +10z = -3

This equation can be written as

1
x 3  5 y  2 z 
10
1
y  3  4 x  3 z 
10
1
z  3  x  6 y 
10

We start with initial values [ x0, y0, z0 ] = (0, 0, 0 )

First Iteration

1
x1  3  5(0)  2(0)   0.3
10
1
y 1 3  4(0)  3(0)   0.3
10
1
z 1 3  0  6(0)   0.3
10

Second Iteration

1
x2  3  5(0.3)  2(0.3)   0.39
10
1
y 2  3  4(0)  3(0)   0.33
10
1
z 2  3  0  6(0)   0.51
10
 x3  0.363; y3  0.303; z3  0.537
Third Iteration
x4  0.344; y4  0.284; z4  0.518
Fourth Iteration
x5  0.338; y5  0.282; z5  0.505
Fifth Iteration
Sixth Iteration x6  0.340; y6  0.284; z6  0.503

Seventh Iteration x7  0.341; y7  0.285; z7  0.504

x8  0.342; y8  0.285; z8  0.505
Eight Iteration
x9  0.342; y9  0.285; z9  0.505
Ninth Iteration

Therefore the solution is x = 0.342; y = 0.285; z = -0.505

EXERCISE 1.5

SOLVE THE FOLLOWING EQUATIONS BY GAUSS JACOBI METHOD:

1. 28x+4y-z =32; x+3y+10z =24; 2x+17y+4z =35

2. 30x-2y+3z =75; 2x+2y+18z =30; x+17y-2z =48.
3. 0x-2y+z =12; x+9y-z =10; 2x-y+11z =20

1.6 GAUSS- SEIDAL ITERATIVE METHOD

Let the given system of simultaneous equations be

a1x + b1y + c1z = d1

a2x + b2y + c2z = d2 …………………………………….. (1)
a3x + b3y + c3z = d3

This system of equations can also be written as

1
x d1  b1 y  c1 z 
a1
1
y d 2 a2 x  c2 z 
b2
1
z d3  a3 x  b3 y 
c1
Substituting x2 =0, x3 =0 in (A), we get the value for x1 and it is denoted by x1,
substituting x1 = x1 and x3 = 0 in (B) we get the value for x2 and it is denoted by x2,
substituting x1 = x1 and x2 = x2 in (C) we get the value for x3. These values of x1,
x2, x3 are called the first iterative value of x1, x2 and x3.

The above process may continue for 2nd , 3rd, 4th etc.,iterations.

EXAMPLE: 6

Solve the following system of equations by using Gauss–Seidel

method 8x – 3y + 2z = 20; 4x + 11y – z = 33; 6x + 3y + 12z = 35.

Solution:
8  3 2 
 
 4 11  1
 6 3 12 
The co-efficient matrix A =   is diagonally dominant.

Solving for x, y, z we have

1
x 20  3 y  2 z ; y  1 33  4 x  z ; z  1 35  6 x  3 y 
8 11 12

Let the initial values be x = 0, y = 0, z = 0

First Iteration:
1
x  20  30  20  2.5
8
1
y  33  42.5  0  2.091
11
1
z  35  62.5  32.091  1.144
12

Second Iteration:
1
x  20  32.091  21.144  2.998
8
1
y  33  42.998  1.144  2.014
11
 1
z 35  62.998  32.014  0.914
12

Third Iteration:
1
x  20  32.014  20.914  3.027
8
1
y  33  43.027   0.914  1.982
11
1
z  35  63.027   31.982   0.908
12

Fourth Iteration:
1
x  20  31.982  20.908  3.016
8
1
y  33  43.016   0.908  1.986
11
1
z  35  63.016   31.986   0.912
12

Fifth Iteration:
1
x  20  31.986  20.912  3.017
8
1
y  33  43.017   0.912  1.986
11
1
z  35  63.017   31.986  0.912
12

Sixth Iteration:
1
x  20  31.986  20.912  3.017
8
1
y  33  43.017   0.912  1.986
11
1
z  35  63.017   31.986  0.912
12

The iteration values are tabulated as follows

Iteration x y z
1 2.500 2.091 1.144
2 2.998 2.014 0.914
3 3.027 1.982 0.908
 4 3.016 1.986 0.912
5 3.017 1.986 0.912
6 3.017 1.986 0.912

Hence, the solution is x = 3.017; y = 1.986; z = 0.912

EXERCISE 1.6

SOLVE THE FOLLOWING EQUATIONS BY GAUSS SEIDEL METHOD:

1. 27x+6y-z = 85; 6x+15y+2z =72; x+y+54z =110
2. 10x-2y+z =12; x+9y-z 10; 2x-y+11z =20
3. 83x+11y-4z =95; 7x+52y+13z =104; 3x+8y+29z =71

1.7 Inverse of a matrix by Gauss Jordan Method

By Gauss-Jordan method, the inverse matrix X is obtained by the following steps:

Step: 1

First consider the augmented matrix (A/I)

Step: 2

Reduce the matrix A in (A/I) to the identity matrix I by employing row

transformations.

The row transformations used in step: 2 transform I to A-1.

Finally write the inverse matrix A-1. So the principle involved for finding A-1 is as
shown below.
Gauss  Jordan
(A/I)    (I/A-1)

Note:

The answer can be checked using the result

A . A-1 = A-1 . A = I

EXAMPLE: 7
 4 1 2
 
2 3  1
 
1 2 2 
By Gauss – Jordan method find A , if A = 
-1
.

Solution:
Consider the augmented matrix
4 1 2 1 0 0
 
 2 3 1 0 1 0
1  2 2 0 0 1 
[A, I] = 
1  2 2 0 0 1
 
 2 3 1 0 1 0
4 1 2 1 0 0  R1  R3
~ 
1  2 2 0 0 1 
 
 0 7  5 0 1  2  R 2  R 2  2 R1
0  5 4 1  2 0  R3  R3  2 R 2
~
1  2 2 0 0 1 
 
 0 1  5 / 7 0 1/ 7  2 / 7  1
0  5 4 1  2 0  R2    R2
~  7
1 0 4 / 7 0 2/7 3/ 7 
 
 0 1  5 / 7 1 1/ 7  2 / 7  R1  R1  2 R2
0 0 3/ 7 0  9 / 7  10 / 7  R3  R3  5 R2
~
1 0 4 / 7 0 2/ 7 3/ 7 
 
0 1  5/ 7 0 1/ 7  2 / 7 
0 0 7
 1 7 / 3  3  10 / 3  R3  R3  3 

4
1 0 0  4 / 3 2 7 / 3  R1  R1   7  R3
 
0 1 0 5/3  2 8/3  5
0 0 1 7 / 3  3  10 / 3  R2  R2    R3
  7
 4/3 2 7/3 
 
A   5/3  2 8/3 
-1

 7 / 3  3  10 / 3 
Hence,  

EXERCISE 1.7
 ( )
1 3 3
1 4 3
1. Find the inverse of A =
1 3 4 by Gauss-Jordan method.

( )
0 1 2
1 2 3
2. Find the inverse of A =
3 1 1 by Gauss- Jordan method.

( )
2 −2 4
2 3 2
3. Find the inverse of A =
−1 1 −1 by Gauss- Jordan method.

1.8 Eigen value of a matrix by Power method

Let λ1, λ2……., λn be the Eigen values of A and let , λ1 be the dominant Eigen
value.
1  2  3  ........  n
i.e.,

If the corresponding Eigen vectors are x0, x1,……xn , then any arbitrary vector y
can be eritten as y = a0x0 + a1x1 + …….+ anxn.
Since the Eigen vectors are linearly independent.

Now,

Ak y  Ak  a0 x0  .......  an xn 
 a01k x0  a12k x1  .....  an nk xn
  
k 
 1k  a0 x0  a1  2  x1  .......

  1  


But i  1, i  2,3,......n 
1
Hence, Ak y  1k a0 x0 and Ak 1 y  1k 1a0 x0

Ak 1 y
1  k
Hence, if k is large, A y where the division is carried out in the
corresponding components.

Here y is quite arbitrary. But generally we choose it as the vector having all its
components one.
Note:

1  2  3  ........  n
1. λ1 is dominant if ., .

2. To find the numerically smallest Eigen value of A, obtain the numerically

greatest Eigen value of A-1 and then take its reciprocal.

3. The Eigen values of A – kI are λi are the Eigen values of A.

4. Sum of the Eigen values of a matrix is equal to the sum of the diagonal
elements of the matrix = Trace of the matrix.

5. To find the numerically smallest Eigen value of A, obtain the dominant

Eigen value λ1 of A and then find B = A - λ1I and find the dominant Eigen
value of B. Then the smallest Eigen value of A is equal to the dominant
Eigen value of B + λ1 .

EXAMPLE: 8

Find the largest Eigen value and the corresponding vector of the matrix A =
 9 10 8 
 
10 5  1
 8 1 3 
 

Solution:

Let X0 =
1 1 1
T
be the initial vector
 9 10 8  1  27   1.00 
      
10 5  1 1 14   0.52 
 8  1 3  1    
AX0 =     = 10  = 27  0.37 

 9 10 8   1.00  17.16   1.00 

      
10 5  1  0.52  12.23   0.71 
 8  1 3   0.37   8.59   
AX1 =    =  = 17.16  0.50 
  9 10 8   1.00   20.10   1.00 
      
10 5  1  0.71   13.05   0.65 
 8  1 3   0.50   8.79   
AX2 =   =   = 20.10  0.44 
 9 10 8   1.00  19.02   1.00 
      
10 5  1  0.65   12.51   0.67 
 8  1 3   0.44   8.67   
AX3 =    =  = 19.02  0.46 
 9 10 8   1.00  19.38   1.00 
      
10 5  1  0.67  12.89   0.66 
 8  1 3   0.46   8.71   
AX4 =    =  = 19.38  0.45 
 9 10 8   1.00  19.20   1.00 
      
10 5  1  0.66  12.85   0.67 
 8  1 3   0.45   8.71   
AX5 =    =  = 19.20  0.45 
 9 10 8   1.00  19.30   1.00 
      
10 5  1  0.67  12.90   0.67 
 8  1 3   0.45   8.68   
AX6 =    =  = 19.30  0.45 
 9 10 8   1.00  19.30   1.00 
      
10 5  1  0.67  12.90   0.67 
 8  1 3   0.45   8.68   
AX7 =    =  = 19.30  0.45 

\ The largest eigen value is 19.3

 1.00 
 
 0.67 
 0.45 
and the corresponding eigen vector is   .

EXERCISE 1.8

1. Use the power method to find the dominant eigen value and the corresponding eigen vector
of the matrix

( )
9 1 8
7 4 1
A=
1 7 9

2. Find the largest Eigen value and the corresponding Eigen vector of the matrix

( )
4 6 0
0 5 3
A=
2 0 3

3. Using power method, find the dominant Eigen value and the corresponding Eigen
 ( )
3 2 6
−1 12 1
vector of the matrix A =
4 2 1