Scribd
Upload
50 views3 pages

Iterative Methods in Numerical Analysis

This document provides 12 numerical methods problems involving Jacobi iteration, Gauss-Seidel iteration, LU decomposition, and the power method. The problems involve finding iterations, solutions, eigenvalues and eigenvectors for various systems of linear equations and matrices. Sample answers are provided for each problem.

Uploaded by

Gaurav Upadhyay
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
50 views3 pages

Iterative Methods in Numerical Analysis

This document provides 12 numerical methods problems involving Jacobi iteration, Gauss-Seidel iteration, LU decomposition, and the power method. The problems involve finding iterations, solutions, eigenvalues and eigenvectors for various systems of linear equations and matrices. Sample answers are provided for each problem.

Uploaded by

Gaurav Upadhyay
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Dr B R AMBEDKAR NIT JALANDHAR

NUMERICAL METHODS (MACI-203)

Tutorial Sheet-II

(1) Find the first two iterations of the Jacobi method for the following linear system, using
X(0) = 0
4x1 + x2 − x3 = 5
−x1 + 3x2 + x3 = −4
2x1 + 2x2 + 5x3 = 1
Answer: X(2) = (1.633333, −.983333, 0.233333)t .

(2) Find the first two iterations of the Jacobi method for the following linear system, using
X(0) = 0
4x1 + x2 − x3 + x4 = −2
x1 + 4x2 − x3 − x4 = −1
−x1 − x2 + 5x3 + x4 =0
x1 − x2 + x3 + 3x4 =1
Answer: X(2) = (−0.5208333, −0.04166667, −0.2166667, 0.4166667)t .

(3) Find the first two iterations of the Gauss-Seidel method for the following linear system,
using X(0) = 0
1
−2x1 + x2 + x3 = 4
2
1
x1 − 2x2 − x3 = −4
2
x2 + 2x3 = 0
Answer: X(2) = (−1.625, 1.3125, −0.65625)t .

(4) Use the Gauss-Seidel iterative technique to find approximate solutions to


10x1 − x2 + 2x3 = 6
−x1 + 11x2 − x3 + 3x4 = 25
2x1 − x2 + 10x3 − x4 = −11
3x2 − x3 + 8x4 = 15
starting with X = (0, 0, 0, 0)t and iterating until
kX (k) − X (k−1) k∞
≤ 10−3 .
kX (k) k∞
Answer: X(5) = (1.0001, 2.0000, −1.0000, 1.0000)t .
(5) The following system of equations are given:

4x1 + x2 + x3 = 4
x1 + 4x2 − 2x3 = 4
3x1 + 2x2 − 4x3 = 6
solve the above system by the LU decomposition method.
Answer: X = [1, 1/2, −1/2]t .

(6) Solve the system of equations


2x1 + x2 − 4x3 + x4 = 4
−4x1 + 3x2 + 5x3 − 2x4 = −10
x1 − x2 + x3 − x4 = 2
x1 + 3x2 − 3x3 + 2x4 = −1
by the LU decomposition method, assuming uii = 1, i = 1, 2, 3, 4.
Answer: X = [1, −1, −1, −1]t .

(7) Find the first three iterations obtained by the Power method applied to the following
matrix.
1 −1 0
" #
−2 4 −2 ;
0 −1 2
Use X(0) = (−1, 2, 1)t .
Answer: µ(3) = 5.000000, X(3) = (−0.2578947, 1, −0.2842105)t

(8) Find the first three iterations obtained by the Power method applied to the following
matrix.  
4 1 1 1
1 3 −1 1
1 −1 2 0 ;
1 1 0 2
Use X(0) = (1, −2, 0, 3)t .
Answer: µ(3) = 5.038462, X(3) = (1, 0.2213741, 0.3893130, 0.4045802)t .

(9) Find inverse of the matrix  


2 1 0 0
1 2 1 0
0 1 2 1
0 0 1 2
by the partition method.
Answer:
4 −3 2 −1
 
1 −3 6 −4 2 
.
5 2 −4 6 −3

−1 2 −3 4

(10) If A and C are non singular matrix then show that


 −1  
A O A−1 O
=
B C −C −1 BA−1 C −1
2
Hence find inverse of  
1 0 0 0
0 2 0 0
3 0 4 0
0 1 0 3
Answer:  
1 0 0 0
 0 1/2 0 0 
−3/4 0 1/4 0  .
0 −1/6 0 1/3

(11) Find all the eigenvalues and the corresponding eigenvectors of the matrix.
 √ 
√1 3 √4
 3 5 3
√
4 3 1
using Jacobi method. √ √ √
Answer: √ Eigen values are 5+
√ 6,√ 5− 6, −3 and the eigenvectors are (1/2)[1, 2, 1]t ,
(1/2)[−1, 2, −1]t , (1/2)[− 2, 0, 2]t .

(12) Find all the eigenvalues of the matrix


3 2 1
" #
2 3 2
1 2 3
using Jacobi method. Iterate till off-diagonal elements in magnitude are less than 0.0005.
Answer: After six rotations, eigen values are 6.372293, 0.627718, 2.000001.

Common questions

Powered by AI

For an invertible block matrix, the inverse is found using the formula: If A and C are nonsingular, then . Applying this to an example matrix results in the inverse being calculated as shown, for instance, in the matrix example from the document .

The power method starts with an initial guess vector and applies the matrix iteratively. For the given matrix, starting with X(0) = (1, -2, 0, 3)t, subsequent iterations provide an approximated dominant eigenvalue of µ(3) = 5.038462 and eigenvector X(3) = (1, 0.221374, 0.389313, 0.404580)t .

Using the Jacobi method, which is an iterative algorithm to diagonalize a symmetric matrix, the eigenvalues are calculated until off-diagonal elements are below a threshold. For this matrix, after six rotations, the eigenvalues are determined as 6.372293, 0.627718, 2.000001 .

Implementation involves solving each equation in sequence using the updated values as soon as they are available. For the given system, the result after two iterations by applying this method provides the estimate X(2) = (-1.625, 1.3125, -0.65625)t .

Iterative methods like Gauss-Seidel require the spectral radius of iteration matrix to be less than 1 for convergence. In the provided system, convergence is achieved after 5 iterations when ∥X(k) − X(k−1)∥∞/∥X(k)∥∞ ≤10⁻³, resulting in the vector X(5) = (1.0001, 2.0000, -1.0000, 1.0000)t .

Jacobi and Gauss-Seidel methods are simple and intuitive for iterative solutions, suitable for large sparse systems. However, their convergence is not guaranteed for every system, dependent on the spectral radius and diagonal dominance conditions. Gauss-Seidel is generally faster due to immediate updates, but both may converge slowly on ill-conditioned systems .

LU decomposition involves decomposing the coefficient matrix into a lower triangular matrix L and an upper triangular matrix U. For the given system, after performing LU decomposition, back substitution leads to the solution X = [1, 0.5, -0.5]t .

The Jacobi method involves solving each equation for its respective variable, using the previous iteration's values. For this system, the first two iterations result in X(2) = (1.633333, −0.983333, 0.233333)t .

The inverse of a 2x2 block matrix is calculated by separately inverting individual submatrices A and C if they are nonsingular, and using a specified formula for the off-diagonal blocks. An example from the document shows the inversion process applied to a specific matrix, generating a precise inverse matrix .

The Gauss-Seidel method differs from the Jacobi method by using the latest updated values of the variables as soon as they are available within the same iteration, rather than only relying on previous iteration values. This often leads to faster convergence compared to the independent updates per equation seen in the Jacobi method .

You might also like