Lab:01

Bisection code in MATLAB

Introduction
Bisection method is used for finding the real root of a given function say f(x)=0. It requires two
starting values x0 and x1 for the solution such that f(x0). f(x1) <0. Then the equation f(x) = 0 has
at least one root in the interval (x0 , x1). Graphically it can be represented as:

Procedure of bisection method rely on repeated application of the following:

1. Find x2, the new iterate using, x2 = (x0 + x1)/2 and evaluate f(x2).
2. If f(x2) = 0 then x2 is a root of f(x).
3. If f(x2) ≠ 0 then two things are possible:
i) If f(x0) f(x2) < 0, we compute the new iterate x3 as:
x3 = (x0 + x2)/2 and evaluate f(x3)
ii) If f(x1) f(x2) < 0, we compute the new iterate x3 as:
x3 = (x1 + x2)/2 and evaluate f(x3)

Matlab code for Bisection Method

A function f(x) = x.^3+4*x.^2-10 is given. Find the root in interval [1 ,2]

Codes for bisection method in matlab:

% Solution of nonlinear equations
% Bisection method
clear all, close all, clc
format short
f = %question;
a=1; b=2;% roots
max_I=20;% maximum number of iterations
tol=10^(-5);% tolerance
fa=feval(f,a); % Initial guess
fb=feval(f,b);
if fa*fb>0,

disp('Wrong choice')
end

for k=1:max_I
c=(a+b)/2;
C(k)=c;
fc=feval(f,c);
Fc(k)=fc;
if fc==0,break
end
if fb*fc>0;
b=c;
fb=fc;
else
a=c;
fa=fc;
end
if abs(fc)<tol,break
end
end

fprintf('no of iteration(k), root sequence(x_k), function values fc\n')

[(1:k);C(1:k);Fc(1:k)]'

Example solve on matlab

% Solution of nonlinear equations
% Bisection method
clear all, close all, clc
format short
f = @(x) 2*x.^3*x.^2+10*x+1;
a=1; b=3;
max_I=15;% maximum number of iterations
tol=10^(-5);
fa=feval(f,a); % Initial guess
fb=feval(f,b);
if fa*fb>0,

disp('Wrong choice')
end

for k=1:max_I
c=(a+b)/2;
C(k)=c;
fc=feval(f,c);
Fc(k)=fc;
if fc==0,break
end
if fb*fc>0;
b=c;
fb=fc;
else
a=c;
fa=fc;
end
if abs(fc)<tol,break
end
end

fprintf('no of iteration(k), root sequence(x_k), function values fc\n')

[(1:k);C(1:k);Fc(1:k)]'

Solution:
Wrong choice
no of iteration(k), root sequence(x_k), function values fc

ans =

1.0000 2.0000 85.0000

2.0000 1.5000 31.1875
3.0000 1.2500 19.6035
4.0000 1.1250 15.8541
 5.0000 1.0625 14.3332
6.0000 1.0312 13.6452
7.0000 1.0156 13.3175
8.0000 1.0078 13.1575
9.0000 1.0039 13.0784
10.0000 1.0020 13.0391
11.0000 1.0010 13.0196
12.0000 1.0005 13.0098
13.0000 1.0002 13.0049
14.0000 1.0001 13.0024
15.0000 1.0001 13.0012

Another example on bisection method solve on matlab:

Codes:
% Solution of nonlinear equations
% Bisection method
clear all, close all, clc
format short
f = @(x) 3*x.^4-x.^3+5*x.^2-7;
a=2; b=5;
max_I=20;% maximum number of iterations
tol=10^(-3);
fa=feval(f,a); % Initial guess
fb=feval(f,b);
if fa*fb>0,
disp('Wrong choice')
end

for k=1:max_I
c=(a+b)/2;
C(k)=c;
fc=feval(f,c);
Fc(k)=fc;
if fc==0,break
end
 if fb*fc>0;
b=c;
fb=fc;
else
a=c;
fa=fc;
end
if abs(fc)<tol,break
end
end

fprintf('no of iteration(k), root sequence(x_k), function values fc\n')

[(1:k);C(1:k);Fc(1:k)]'

Solution:

Wrong choice
no of iteration(k), root sequence(x_k), function values fc

ans =

1.0000 3.5000 461.5625

2.0000 2.7500 181.5898
3.0000 2.3750 103.2566
4.0000 2.1875 75.1514
5.0000 2.0938 63.3932
6.0000 2.0469 58.0334
7.0000 2.0234 55.4768
8.0000 2.0117 54.2285
9.0000 2.0059 53.6118
10.0000 2.0029 53.3053
11.0000 2.0015 53.1525
12.0000 2.0007 53.0762
13.0000 2.0004 53.0381
14.0000 2.0002 53.0190
15.0000 2.0001 53.0095
16.0000 2.0000 53.0048
17.0000 2.0000 53.0024
18.0000 2.0000 53.0012
 19.0000 2.0000 53.0006
20.0000 2.0000 53.0003

Lab: 02

Title: Newton Raphson method.

Introduction:
In numerical analysis, Newton's method, also known as the Newton–Raphson method, named
after Isaac Newton and Joseph Raphson, is a root-finding algorithm which produces successively
better approximations to the roots (or zeroes) of a real-valued function. The most basic version
starts with a single-variable function f defined for a real variable x, the function's derivative f ′,
and an initial guess x0 for a root of f. If the function satisfies sufficient assumptions and the
initial guess is close, then
f (x 0)
X 1 = x0 +
f ( x 0)'
is a better approximation of the root than x0. Geometrically, (x1, 0) is the intersection of the x-
axis and the tangent of the graph of f at (x0, f (x0)): that is, the improved guess is the unique root
of the linear approximation at the initial point. The process is repeated as
f ( xn)
xn+1 = xn -
f ( xn ) '
until a sufficiently precise value is reached. This algorithm is first in the class of Householder's
methods, succeeded by Halley's method. The method can also be extended to complex
functions and to systems of equations.

Codes for this method in matlab:

% Solution of nonlinear equations

% Newton-Raphson method
clear all, close all, clc
format short
f = @(x) x.^3+4*x.^2-10;
df= @(x) 3*x.^2+8*x;
N=20;% maximum number of iterations
tol=10^(-3);
x0=2; % Initial guess
x(1)=x0;
for k=2:N
x(k) = x(k-1)-feval(f,x(k-1))/feval(df,x(k-1));
error=abs(x(k)-x(k-1));
if (error<tol), break;
end
end
% fprintf('no of iteration(k), root sequence(x_k) \n')
[(1:k);x(1:k)]'

Example on this method in matlab:

% Solution of nonlinear equations

% Newton-Raphson method
clear all, close all, clc
format short
f = @(x) 2*x.^3+x.^2-10;
df= @(x) 6*x.^2+2*x;
N=20;% maximum number of iterations
tol=10^(-3);
x0=2; % Initial guess
x(1)=x0;
for k=2:N
x(k) = x(k-1)-feval(f,x(k-1))/feval(df,x(k-1));
error=abs(x(k)-x(k-1));
if (error<tol), break;
end
end
% fprintf('no of iteration(k), root sequence(x_k) \n')
[(1:k);x(1:k)]'

Solution in matlab:

ans =

1.0000 2.0000
2.0000 1.6429
3.0000 1.5624
4.0000 1.5585
5.0000 1.5585

 
Other example:

% Solution of nonlinear equations

% Newton-Raphson method
clear all, close all, clc
format short
f = @(x) x.^3-x.*sin(x);
df= @(x) 3*x.^2-x.*cos(x)+sin(x);
N=20;% maximum number of iterations
tol=10^(-3);
x0=2; % Initial guess
x(1)=x0;
for k=2:N
x(k) = x(k-1)-feval(f,x(k-1))/feval(df,x(k-1));
error=abs(x(k)-x(k-1));
if (error<tol), break;
end
end
% fprintf('no of iteration(k), root sequence(x_k) \n')
[(1:k);x(1:k)]'

SOLUTION:

ans =

1.0000 2.0000
2.0000 1.5502
3.0000 1.2841
4.0000 1.1243
5.0000 1.0276
6.0000 0.9688
7.0000 0.9329
8.0000 0.9111
9.0000 0.8977
10.0000 0.8896
11.0000 0.8846
12.0000 0.8815
13.0000 0.8797
14.0000 0.8785
15.0000 0.8778
 
LAB: 03

TITLE: REGULA FALSI METHOD

INTRODUCTION:
n mathematics, the regula falsi, method of false position, or false position method is a very old
method for solving an equation with one unknown, that, in modified form, is still in use. In
simple terms, the method is the trial and error technique of using test ("false") values for the
variable and then adjusting the test value according to the outcome. This is sometimes also
referred to as "guess and check". Versions of the method predate the advent of algebra and the
use of equations.

The method of false position provides an exact solution for linear functions, but more direct
algebraic techniques have supplanted its use for these functions. However, in numerical analysis,
double false position became a root-finding algorithm used in iterative numerical approximation
techniques.
Many equations, including most of the more complicated ones, can be solved only by iterative
numerical approximation. This consists of trial and error, in which various values of the
unknown quantity are tried. That trial-and-error may be guided by calculating, at each step of the
procedure, a new estimate for the solution. There are many ways to arrive at a calculated-
estimate and regula falsi provides one of these.

CODES OR REGULA FALSI METHOD:

% Solution of nonlinear equations
% Regula-Falsi
clear all,close all
f = @(x) x.^3+4*x.^2-10;
a=1; b=2;
N=20;% maximum number of iterations
tol=10^(-5);
fa=feval(f,a); % Initial guess
fb=feval(f,b);
if fa*fb>0
%break
disp('Wrong choice')
end

for k=1:N
c=b-(b-a)*fb/(fb-fa);
C(k)=c;
fc=feval(f,c);
Fc(k)=fc;
if fc==0,break
 end
if fb*fc>0;
b=c;
fb=fc;
else
a=c;
fa=fc;
end
if abs(fc)<tol,break
end
end

fprintf('no of iteration(k), root sequence(x_k), function values fc\n')

[(1:k);C(1:k);Fc(1:k)]'

EXAMPLE ON REGULA FALSI METHOD:

% Solution of nonlinear equations
% Regula-Falsi
clear all,close all
f = @(x) 2*x.^4+6*x.^3+x.^2-5*x-10;
a=4; b=7;
N=20;% maximum number of iterations
tol=10^(-5);
fa=feval(f,a); % Initial guess
fb=feval(f,b);
if fa*fb>0
%break
disp('Wrong choice')
end

for k=1:N
c=b-(b-a)*fb/(fb-fa);
C(k)=c;
fc=feval(f,c);
Fc(k)=fc;
if fc==0,break
end
if fb*fc>0;
b=c;
fb=fc;
else
a=c;
fa=fc;
end
if abs(fc)<tol,break
end
end

fprintf('no of iteration(k), root sequence(x_k), function values fc\n')

[(1:k);C(1:k);Fc(1:k)]'

SOLUTION:
untitled10
Wrong choice
no of iteration(k), root sequence(x_k), function values fc

ans =

1.0000 3.5577 575.4475

2.0000 2.7274 216.1874
3.0000 2.3141 115.4969
4.0000 2.0601 72.4260
5.0000 1.8866 49.7474
6.0000 1.7602 36.2208
7.0000 1.6643 27.4534
8.0000 1.5893 21.4236
9.0000 1.5293 17.0889
10.0000 1.4804 13.8648
11.0000 1.4402 11.4011
12.0000 1.4067 9.4770
13.0000 1.3785 7.9475
14.0000 1.3547 6.7135
15.0000 1.3344 5.7056
16.0000 1.3170 4.8737
17.0000 1.3021 4.1812
18.0000 1.2893 3.6002
19.0000 1.2782 3.1096
20.0000 1.2685 2.6931

ANOTHER EXAMPLE:
% Solution of nonlinear equations
% Regula-Falsi
clear all,close all
f = @(x) x*sin(x)-1;
a=0; b=2;
N=20;% maximum number of iterations
tol=10^(-5);
fa=feval(f,a); % Initial guess
fb=feval(f,b);
if fa*fb>0
%break
disp('Wrong choice')
end

for k=1:N
c=b-(b-a)*fb/(fb-fa);
C(k)=c;
fc=feval(f,c);
Fc(k)=fc;
if fc==0,break
end
if fb*fc>0;
b=c;
fb=fc;
else
a=c;
fa=fc;
end
if abs(fc)<tol,break
end
end

fprintf('no of iteration(k), root sequence(x_k), function values fc\n')

[(1:k);C(1:k);Fc(1:k)]'

SOLUTION:
untitled11
no of iteration(k), root sequence(x_k), function values fc

ans =

1.0000 1.0998 -0.0200

2.0000 1.1212 0.0098
3.0000 1.1142 0.0000

 
LAB:04

TTLE: SECANT METHOD:

INTRODUCTON:
In numerical analysis, the secant method is a root-finding algorithm that uses a succession
of roots of secant lines to better approximate a root of a function f. The secant method can be
thought of as a finite-difference approximation of Newton's method.

The secant method is very similar to the bisection method except instead of dividing each
interval by choosing the midpoint the secant method divides each interval by the secant line
connecting the endpoints.

CODES FOR SECANT METHOD

% Solution of nonlinear equations
%Secant method method

clear all,close all

f = @(x) sqrt(x)-cos(x);
format short
N=20;% maximum number of iterations
tol=10^(-5);
% Initial guess
x(1)=1.0;x(2)=0.5;
for k=3:N
x(k) = x(k-1)-feval(f,x(k-1))*(x(k-1)-x(k-2))/(feval(f,x(k-1))-feval(f,x(k-2)));
error=abs(x(k)-x(k-1));
if (error<tol), break;
end
end
fprintf('no of iteration(k), root sequence(x_k) \n')
[(1:k);x(1:k)]'

EXAMPLE ON THIS METHOD IN MATLAB:

% Solution of nonlinear equations
%Secant method method
clear all,close all
f = @(x) sqrt(x)-sin(x);
format short
N=20;% maximum number of iterations
tol=10^(-5);
% Initial guess
x(1)=1.0;x(2)=0.5;
for k=3:N
x(k) = x(k-1)-feval(f,x(k-1))*(x(k-1)-x(k-2))/(feval(f,x(k-1))-feval(f,x(k-2)));
error=abs(x(k)-x(k-1));
if (error<tol), break;
end
end
fprintf('no of iteration(k), root sequence(x_k) \n')
[(1:k);x(1:k)]'

SOLLUTION:
untitled12
no of iteration(k), root sequence(x_k)

ans =

1.0000 + 0.0000i 1.0000 + 0.0000i

2.0000 + 0.0000i 0.5000 + 0.0000i
3.0000 + 0.0000i 2.1462 + 0.0000i
4.0000 + 0.0000i -0.4409 + 0.0000i
5.0000 + 0.0000i 1.4746 + 2.2377i
6.0000 + 0.0000i 0.1607 + 0.0761i
7.0000 + 0.0000i 0.2288 + 0.2424i
8.0000 + 0.0000i 0.3421 - 0.7550i
9.0000 + 0.0000i 1.2073 + 0.4307i
10.0000 + 0.0000i 1.5804 + 0.6270i
11.0000 + 0.0000i 1.0720 + 0.5687i
12.0000 + 0.0000i 1.0906 + 0.6594i
13.0000 + 0.0000i 1.1391 + 0.6473i
14.0000 + 0.0000i 1.1354 + 0.6508i
15.0000 + 0.0000i 1.1352 + 0.6506i
16.0000 + 0.0000i 1.1352 + 0.6506i

 
ANOTHER EXAMPLE:
% Solution of nonlinear equations
%Secant method method

clear all,close all

f = @(x) x-0.25*sin(x)-0.5;
format short
N=20;% maximum number of iterations
tol=10^(-5);
% Initial guess
x(1)=0.5;x(2)=0.8;
for k=3:N
x(k) = x(k-1)-feval(f,x(k-1))*(x(k-1)-x(k-2))/(feval(f,x(k-1))-feval(f,x(k-2)));
error=abs(x(k)-x(k-1));
if (error<tol), break;
end
end
fprintf('no of iteration(k), root sequence(x_k) \n')
[(1:k);x(1:k)]'

SOLUTION:
untitled13
no of iteration(k), root sequence(x_k)

ans =

1.0000 0.5000
2.0000 0.8000
3.0000 0.6495
4.0000 0.6516
5.0000 0.6516
6.0000 0.6516

 
 

LAB:05

TITLE: GUASS ELIMATION METHOD

INTRODUCTION:
Gaussian elimination, also known as row reduction, is an algorithm in linear algebra for solving
a system of linear equations. It is usually understood as a sequence of operations performed on
the corresponding matrix of coefficients. This method can also be used to find the rank of a
matrix, to calculate the determinant of a matrix, and to calculate the inverse of an invertible
square matrix. 

To perform row reduction on a matrix, one uses a sequence of elementary row operations to
modify the matrix until the lower left-hand corner of the matrix is filled with zeros, as much as
possible. There are three types of elementary row operations:

1. Swapping two rows,

2. Multiplying a row by a nonzero number,
3. Adding a multiple of one row to another row.

Using these operations, a matrix can always be transformed into an upper triangular matrix,

CODES FOR ELIMINATION METHOD:

clear all, close all, clc
% Gauss Elimination method
A=[4 1 2; 2 4 -1; 1 1 -3] % is N*N matrix
B=[9;-5;-9] % is N*1 martix
[N N]= size(A);
X=zeros(N,1);
C=zeros(1, N+1);
Aug= [A B]; % augmented matrix
for p=1:N-1
[Y,j]=max(abs(Aug(p:N,p))); %partial pivoting for column p
% interchange row p and j
C=Aug(p,:);
Aug(p,:)=Aug(j+p-1,:);
Aug(j+p-1,:)=C;
if Aug(p,p)==0, break
end
% elimination of column p
for k=p+1:N
m=Aug(k,p)/Aug(p,p);
Aug(k,p:N+1)=Aug(k,p:N+1)-m*Aug(p,p:N+1);
end
end
% Back substitution
X(N)=Aug(N,N+1)/Aug(N,N);
for k=N-1:-1:1
X(k)=(Aug(k,N+1)-Aug(k,k+1:N)*X(k+1:N))/Aug(k,k);
end

EXAMPLE ON ELIMINATION METHOD IN MATLAB

clear all, close all, clc
% Gauss Elimination method
A=[4 1 2; 2 4 -1; 1 1 -3] % is N*N matrix
B=[9;-5;-9] % is N*1 martix
[N N]= size(A);
X=zeros(N,1);
C=zeros(1, N+1);
Aug= [A B]; % augmented matrix
for p=1:N-1
[Y,j]=max(abs(Aug(p:N,p))); %partial pivoting for column p
% interchange row p and j
C=Aug(p,:);
Aug(p,:)=Aug(j+p-1,:);
Aug(j+p-1,:)=C;
if Aug(p,p)==0, break
end
% elimination of column p
for k=p+1:N
m=Aug(k,p)/Aug(p,p);
Aug(k,p:N+1)=Aug(k,p:N+1)-m*Aug(p,p:N+1);
end
end
% Back substitution
X(N)=Aug(N,N+1)/Aug(N,N);
for k=N-1:-1:1
X(k)=(Aug(k,N+1)-Aug(k,k+1:N)*X(k+1:N))/Aug(k,k);
end

SOLUTION:

A=

4 1 2
2 4 -1
1 1 -3

B=

9
 -5
-9

 
ANOTHER EXAMPLE:
clear all, close all, clc
% Gauss Elimination method
A=[1 2; 2 7; 1 1] % is N*N matrix
B=[9 6;-5 1;-9 9] % is N*N martix
[N N]= size(A);
X=zeros(N,1);
C=zeros(1, N+1);
Aug= [A B]; % augmented matrix
for p=1:N-1
[Y,j]=max(abs(Aug(p:N,p))); %partial pivoting for column p
% interchange row p and j
C=Aug(p,:);
Aug(p,:)=Aug(j+p-1,:);
Aug(j+p-1,:)=C;
if Aug(p,p)==0, break
end
% elimination of column p
for k=p+1:N
m=Aug(k,p)/Aug(p,p);
Aug(k,p:N+1)=Aug(k,p:N+1)-m*Aug(p,p:N+1);
end
end
% Back substitution
X(N)=Aug(N,N+1)/Aug(N,N);
for k=N-1:-1:1
X(k)=(Aug(k,N+1)-Aug(k,k+1:N)*X(k+1:N))/Aug(k,k);
end

SOLUTION:

A=

1 2
2 7
1 1

B=
 9 6
-5 1
-9 9

LAB:06

TITLE: TRAPIZIODAL RULE

INTRODUCTION:
n mathematics, and more specifically in numerical analysis, the trapezoidal rule (also known as
the trapezoid rule or trapezium rule—see Trapezoid for more information on terminology) is a
technique for approximating the definite integral.
b

∫ f ( x ) dx
a

The trapezoidal rule works by approximating the region under the graph of the function f(x) as
a trapezoid and calculating its area. It follows that
b
f ( a ) + f (b)
.∫ f ( x ) dx=( b−a ) .
a 3

CODES FOR THE TRAPIZIADAL RULE:

% Composite Trapezoidal rule
a=0;b=1;n=6;
f =@(x) x.^3;
h=(b-a)/n;
s=0;
for k=1:n-1
x=a+h*k;

s=s+feval(f,x);
end

s=h*(feval(f,a)+feval(f,b))/2+h*s;
s
EXAMPLE ON THIS METHOD:
% Composite Trapezoidal rule
a=0;b=3;n=6;
f =@(x) x.^2 / 1+x.^3 ;
h=(b-a)/n;
s=0;
for k=1:n-1
x=a+h*k;

s=s+feval(f,x);
end

s=h*(feval(f,a)+feval(f,b))/2+h*s;
s

SOLUTION:

s=

29.9375

 
ANOTHER EXAMPLE:
% Composite Trapezoidal rule
a=3;b=7;n=6;
f =@(x) 1-x.^3 /4 ;
h=(b-a)/n;
s=0;
for k=1:n-1
x=a+h*k;

s=s+feval(f,x);
end

s=h*(feval(f,a)+feval(f,b))/2+h*s;
s

SOLUTION:
s=

-142.1111

LAB:07

TITLE: SIMPSON METHOD

INTRODUCTION:
In numerical integration, Simpson's rules are
several approximations for definite integrals, named after Thomas Simpson (1710–1761).
The most basic of these rules, called Simpson's 1/3 rule, or just Simpson's rule, reads
b
a+b
∫ f ( x ) = b−a
6
¿ ¿f(a) + 4f
2
+ f (b)
a

if the 1/3 rule is applied to n equal subdivisions of the integration range [a,b],
one obtains the composite Simpson's rule. Points inside the integration range are given
alternating weights 4/3 and 2/3.
Simpson's 3/8 rule, also called Simpson's second rule requests one more function evaluation
inside the integration range, and is exact if f is a polynomial up to cubic degree.

CODES FOR SIMPSON RULE:

a=-0.25; b=0.25; n=4;
f =@(x) (cos(x)).^2;
h=(b-a)/n;
s1=0;
s2=0;
for k=1:n
x=a+h*(2*k-1);
s1=s1+feval(f,x);
end
for k=1:n-1
x=a+h*2*k;
s2=s2+feval(f,x);
end
s=h/3*(feval(f,a)+feval(f,b)+4*s1+2*s2);
s
EXAMLE ON THIS METHOD:
a=-0; b=90; n=6;
f =@(x) x.^2/1+x.^3;
h=(b-a)/n;
s1=0;
s2=0;
for k=1:n
x=a+h*(2*k-1);
s1=s1+feval(f,x);
end
for k=1:n-1
x=a+h*2*k;
s2=s2+feval(f,x);
end
s=h/3*(feval(f,a)+feval(f,b)+4*s1+2*s2);
s

SOLUTION:
untitled7

s=

374.4000

ANOTHER EXAMPLE:
a=-0; b=90; n=6;
f =@(x) sin(x)^0.5;
h=(b-a)/n;
s1=0;
s2=0;
for k=1:n
x=a+h*(2*k-1);
s1=s1+feval(f,x);
end
for k=1:n-1
x=a+h*2*k;
s2=s2+feval(f,x);
end
s=h/3*(feval(f,a)+feval(f,b)+4*s1+2*s2);
s

SOLUTION:
s=

82.3028 +56.0736i