1.

State whether the following statements are

true or false. Give a short proof or a
counter-example in support of your answer. 10x2=20

(a) The equation x3 — 4x — 16 =0 has no root

in the interval [3, 5].

(b) AE = EA, where E is the shift operator and

A is the forward difference operator.

(¢c) Every 3 x 3 system of linear equations can be

solved using the LU decomposition method.

(d) For the data (2, 4), (1, 5), (3, 6) the Newton’s
divided difference f[x, x;, xgl is g .

(e) The Newton-Raphson method cannot be

used to find a cube root of a positive real
number.
1 2]
(f) The eigenvalues of the matrix ‘ are
0o 3
1and 2.

(g) The sum of all the Lagrange’s fundamental

polynomials is a constant.

(h) If the IVP g—y =2xy, y(0)=1 is solved by

x
the optimal R-K method of second order with
h=01, then K; and Ky are 0 and %,
respectively.
3
(i) The equation i:x%er%er:O is a
linear differential equation.

(G) For any data {(x;,£)]i=0,1,2,3},

A3fy = f3 - 3fy + 3f; — £,

2. (a) Show that the system of linear equations

X1+ 2x3=3
3xg=2
4xq + 2x9 + bxg = 11

is consistent. Hence, find its solution using

the LU decomposition method. 4

(b) Using Horner’s method find P(~1) and P'(-1),

where P(x) = 2x* + 3x% — 4x% + 6x — 10. 3
(0 Perform three iterations of Newton-Raphson
method to approximate a root of the
equation fx) = x* —x + 3 = 0. You may take
X = 0 as the initial guess.

(a) Estimate the value of f(2) from the

following data using Lagrange’s
interpolation :

X -1 0 1 3
f(x) 0|-4] 0 56

(b) Show by the principle of mathematical

induction that A"e* = (e — 1)"e¥, where
Ais the forward difference operator.

The equation x2 + ax + b = 0 has two real

roots o and B such that |a|<|B|. If we use

the fixed point iteration x,,; = to
Xp+a
find a root, then to which root does it

converge ?

(a) For the data

Xj -3|-2|-1| 0 11213

fi B 7] 3|1|1]|3]|7

show that A3=0 Vi=0,1,2,3.

(b) Using the Taylor series method of second

order, find the approximate value of y(1-4)
for the IVP y’ = x2 + y2, y(1) = 1. Take the
step size h = 0-2.

(a) Obtain the approximate value of

2
X 2dx using Simpson’s rule with
1+x
0
3 and 5 nodal points. Obtain the improved
value using the Romberg integration.

(b) The equation x 3 —x—1=0 has aroot

in the
interval ]1,2[. Determine a suitable
iteration function g(x) so that the iteration
method x,; = g(x,), n =0, 1, 2, ... converges
to the root. Perform two iterations of this
method with xp = 16.
(a) Perform three iterations of the inverse
power method to obtain an eigenvalue of
the following matrix A, which is nearest to
3 in magnitude :
2 -1 0|
A=|-1 2 -1
0o -1 2
Assume the initial approximation to the
eigenvector as (1, —1, T

52
(b) Prove that p2=1+ e

(a) Using an appropriate interpolation formula,

find the value of f(1:32) from the following
table of values :

x fix)
11 1:3357
12 1-5095

1-3 1-6984

14 1-9043

15 2:1293

(b) Determine the spacing h in a table of equally

spaced points between 2 and 5 that yields a

five-place accuracy in the interpolation of

the function flx) = x+ % by a second

degree polynomial.

(a) Show that the interpolating polynomial of

degree at most n with nodes xq, Xy, ..., X,
can be written as

ofx)
Palo = X)Z (x - x3) Wixg)

where w(x) = (x = Xg) (x = x) ... (x = Xp).

(b) Find the inverse of the matrix

3 01 2
A=[2 -3 -1
1 -2 1l
using the Gauss-Jordan method.
9 a) Solve the system of linear equations
4x) +Xg+xg3=1
X1 +4x9 —2x3 =4
—Xq + 2X9 — 2xg3 = 2
using the Gauss elimination method.

(b) Use the Bisection method to find a root of

the equation x®—4x +1=0. Starting with
the interval [0, 1], stop the iterations if the
interval width becomes smaller than 0-05.

10. (a) Find an approximate value of 326 using

the Mean Value Theorem.

(b) Without computing the eigenvalues of the

matrix
6 2 2|
A=(2 4 1
2 1 3]

prove that the eigenvalues satisfy the

inequality 0 <A < 10.
1. Which of the following statements are true and
which are false ? Give a short proof or a counter-
example in support of your answer : 2x10=20
(@) AV=A-V

(®) If
A= l-a -a I
a 1l-a]

then A" =nA+ (1-n) L

© Every square matrix is invertible.

[CY) A =1]2 2]
2] then for -23 < k<0, the
spectral radius of the matrix (I + kA) is less
than 1.
(e) If :

Fx+2h)—2f (x+h)+f(x) = h2f" (&)

for some & € Jx,x + 2h[, then h > 0.
1 1
® 5=E2+E 2
(®) All the eigen values of the matrix :
1 -1 2]
0 1 3
4 2 1)
lie in the interval [0, 6].

(h) If :
P(x) = 2x* + 3x% - 4x? + 6x - 10,

then P'(-1) = 15.

@ The inverse of a upper triangular matrix is

a lower triangular matrix.
(U] The maximum number of real (positive or
negative) roots of the equation :

ax* +4x% +3x* —x-1=0

is 4.
 (@) Find an interval of unit length that

contains the smallest positive root of the

equation x? —x-10 = 0. Take the end

points of this interval as initial

approximation and perform two iterations

of the Regula-Falsi method. 5

(b) Show by induction that

Am (ex) = (e" - l)n e*, where A is the

forward difference operator. 3

© Show that : 2

A= (i + fin)s
where A is the forward difference operator

and f; = f(x;).

(a) Find the inverse of the matrix : 5

1 2 3]
2 34
-11 1))

using LU factorization method.

(b) Derive the Newton-Raphson method to

obtain a simple root of the equation
f(x) = 0. Find its rate of convergence. 5

4. (a) A linear system of equations Ax =b,

10 k]
where A={2 1 3|, k=1 is to be
k0 1)
solved using the Gauss-Jacobi iteration
method. Find the condition on k so that
this method converges. 5

(b) The function f(x)=(1+ ac)fi is to be

tabulated at equispaced points in the

interval [0, 1] using quadratic
interpolation. Find the largest step size
that can be used so that the error
< 5x%107% in magnitude. 5
 (@) Perform three iterations of the inverse
power method to obtain the smallest eigen

1]
value in magnitude of the matrix [ 3 5\ .

Take the initial approximation to the eigen

vector as [0.9, 1.0]T. 5

(b) Obtain the Lagrange’s interpolating

polynomial which fits the following data :

x )
-1 1

1 3
5 31
7 57

Hence find f(3). 5

6. (a) Obtain an approximate value of y(1.4)

using the Taylor’s series method of order
three, for the initial value problem
Y =x-y%y(1)=2withh=02. 5
() Evaluate the integral :

1 x
L[szfldx

using composite Simpson’s rule with 3 and

5 nodal points. Find the improved value
using Romberg integration. 5

(a) For the numerical differentiation formula :

F(x0) = (30 + 1)~ F(x0 - )]

1

(1) Find the order of the method. 2

(i) Write the error term in power series

of h. 2
(iii) Derive the corresponding Richardson’s
extrapolation scheme. 2
 () The equation x* -5.25 x2 - 6.25 = 0 has
a root close to — 2.2. Perform one iteration
of the Birge-Vieta method to find the root.
4

8. (a) The equation x? + ax + b = 0 has two real

roots p and g such that ’p| < ‘q| If we use

the fixed point iteration Xpyy = — to

X, ta
find a root, then to which root does it
converge ? 5
(b) Find a bound on the eigen values of the
matrix :

3 2))
using Gerschgorin bounds. 5
9. (a) Using the Runge-Kutta fourth order
method with A = 0.2, find an approximate
value of y (0.2) for the initial value problem
¥ =x?+y%y(0)=1. 5

(®) Using the Gauss-Jordan method, find the

inverse of the following matrix : 5
1 1 3]
1 3 -3
S —
10. (a) From the following data, find the number
of students having weight between 60 and
70 kg : 5
Weight (in kg) No. of Students

0—40 250
40—60 120
60—80 100
80—100 70
100—120 50

() Construct a fixed point iteration form

x= g(x) for the equation x3 + x2 -1=0
so that the method converges in the
interval [0, 1]. 5
1. State whether the following statements are
true or false. Give a short proof or a counter-
example in support of your answer : 10

8 Avg=s-X
vV A

(i1) For the function f(x)= l, the second order

X

divided difference based on the points

Xg» X5 Xy 18 f 2, %1, X5] = 25 % x5 .
(1) In the Taylor series expansion for the

function f(x)=x'2 in [-1, 1] about x=0,

the bounds of the error R;(x) exists.

(iv) The interval [ 1, 1] contains all the eigen

1 2 3
values of the matrix [2 -3 4.
3 4 2

The order of convergence of Newton-

Raphson method is 2.

Find an interval of unit length which

contains the smallest positive root of the

equation f(x)=x3-5x+1=0. Taking the

end points of this interval as initial
approximations, perform two iterations of
the secant method. 8

(b) Evaluate the integral j using

0 3+ x

Trapezoidal rule with 2 and 4 subintervals.

Determine the minimum number of
subintervals required if the error in
magnitude is less than 0.002. 7
(a) Perform three iterations of the Gauss-
Seidel iteration method to solve the system
of equations : 8

4x+2z=6

5y+2z=-3

5x+4y+10z=11

Take the initial approximation as

(1.5,-0.6,1.1)T.
(b) Using divided difference, show that the
data : 7

x f(x)

-3 18

-2 12

-1 8

1 6

2 8

3 12

represent a second degree interpolating

polynomial. Hence, obtain the polynomial.
(a) Find an approximate value of y (1.2) for the
initial value problem : 8

gy =x2+2y%, y()=1,
dx

using the classical Runge-Kutta fourth

order method with A =0.1.

(b) Prove that :

1
= f1+—82 ,
}l 4

where & is the central difference operator

and p is the mean operator. 7

(a) Using Taylor series third order method,

solve the initial value problem : 8

Ddx xsy?, yO)=1

upto x= 0.4 with h = 0.2.

(b) The solution of the system of equations

1 4||x|_|[5] .
[6 2] [y]_[S] is attempted by the

Gauss-Jacobi and Gauss-Seidel iteration

scheme. Set up the two schemes in the
matrix form. Will the iteration scheme
converge ? Justify your answer. 7
6. (a) Solve the system of equations : 8
dx+y+z=4

x+4y-2z=4

3x+2y—-4z=6

using LU factorization method. Take the

diagonal elements of U as 1.

(b) The equation f(x)=18x%-33x2+2x+5=0

has 3 real roots. Find the intervals which

contain each of these roots. Perform two

iterations of the bisection method to obtain

the negative root. 7

(a) Find the largest eigen value in magnitude

of the matrix : 8

2 1 1
A=[1 1 0
1 0 2

using the power method. Take the initial

approximation to the eigen vector as

v©@=[111]" and perform four iterations.

(b) Using Lagrange interpolation and the
following data : 7

x EY)
-3 -29
-1 -1
0 1
2 11

find the approximate value of f (1).

(a) Solve the system of equations : 8

x+2y+3z=5
2x+8y+222=6
3x+22y+82z=-10

using Gauss elimination method with

pivoting.
(b) The method :

1{5'1
X n+l =g
5N N 2
% +———]
2 ,n=0, 1,2,...
5
xn xn

where N is a positive constant, converges

to N3, Find the rate of convergence of the
method. 7
1. State whether the following statements are

true or false ? Give a short proof or a counter-

example in support of your answer : 2x5=10

@ For decreasing the number of iterations in

Newton-Raphson method, the value of

['(x) must be increased.

(i) If the bisection method is applied to

compute a root of the equation :

flx)=x*-2"~x?-4=0

in the interval [1, 9], the method converges

to a solution after 3 iterations.

(iii) In Jacobi’s method, the coefficient matrix

has no zeroes on its main diagonal.

@iv) A=E+1

W) Newton forward difference method

can be used only for unequally spaced

intervals.

(a) Apply classical Runge-Kutta fourth order

method to find an approximate value of y

when x = 1.2, given that ji:xy, where

x

¥(1) =2 with h=0.2. 8

(b) Perform three iterations of the Newton-

Raphson method to obtain the aproximate

value of (17)'® starting with initial

approximation X, =2. 7
3. (a) Given that :

7(0)=1, f(1)=3, f(3)=55

find the unique polynomial of degree 2 or

less, which fits the given data. 6

(b) Jacobi iteration method is used to solve the

system of equations :

4 0 3[x] [05]
0 3 2|x,| =|0.5
3 2 5|l [0
Determine the rate of convergence of the
method. 6

(© Does a bound for the error Rs (¥) in the

Taylor's series expansion for the function
5
f(x)=x2 in [-1, 1] about x = 0 exist ?
Justify your answer. 3

4. (a) For the following data interpolate at

x = 025 using forward difference
polynomial : 8

* f(x)
0.1 1.40

0.2 1.56
0.3 1.76
0.4 2.00

0.5 2.28

(b) Determine the order of convergence of the

iterative method :

ol ()5 (51)
" () ()
for finding a simple root of the equation
(a) Estimate the eigenvalues of the

matrix :

using Gerschgorin bound. Draw a rough

sketch of the region where the eigen values

lie. 8

(®) A particle is moving along a straight line.

The displacements x of the particle at some

time instance ¢ are given below :

t x

0 5

1 8

2 12

3 17

4 26

Find the velocity and acceleration of the

particle at ¢ = 4. 7

(a) Solve the following system by the method

of LU decomposition : 8

2x+3y+z=9

x+2y+3z=6

Bx+y+2z=8

Take : hy=lyy=lz=1.
 ®) Apply Lagrange’s method to the given data

to find the value of x when f(x) =15 : 7

x /(%)
5 12
6 13
9 14
1 16
(a) Solve the following system of linear
equations Ax=b with partial pivoting
using Guass-Elimination method : 8

2% +3xy
- x3=5

4x) +4xy —3x3 =3

=20 +3xy —x3 =1

(b) If :

1+p%8% = C, + Cyd? + Cy3*

where u is the mean operator and § is the

central difference operator, find the values

of C;,C, and Cs. 7

() The equation x%—x=1=0 has a root in the

interval [1, 2]. Determine a suitable
iteration function g(x) so that the

iteration method x, = g(x;,), where & =0,

1, 2, converges to the root. Perform

two iterations of this method with x; =1.6.

(b) Obtain the approximate value of

J.Z x 7 dx using Simpson’s rule with 3

01+x
and 5 nodal points. Obtain the improved
value using Romberg integration. 7
Which of the following statements are true and
which are false ? Give a short proof or a
counter- example in support of your answer :
2x10=20
(i) The equation x3-4x-16=0 has a root in
the interval [3, 4].

P.T.O.

121 BMTE-144

(i) The order of convergence of the secant

method is 0.62.

L 1
(if)) p=F2+E 2.
(iv) For the system of linear equations :
Sx+y+2:=34
4y-3z=12
10x-2y+z=-4
the matrix is diagonally dominant.
(v) If h is the step length, then (A-V)x2 is
equal to 2h.
(vi) x=1 is a fixed point for the function
[@=x+1.
(vii) The numerical method :
22h*
ym=[1+m+ ]yn, 1>0, n>0

is relatively stable.
(viii) The composite trapezoidal rule for
evaluation of the integral f/(x)dx
requires the interval [a,b] to be divided
into any number of subintervals of equal
width.

(ix) The method :

converges to 1.5 for any choice of initial

approximation.
(x) The error of any quantity is given as the
absolute value of the difference between
the true value and approximate value of
the quantity.
2. (a) Using Newton-Raphson method, find an
iterative formula to compute the reciprocal

of a natural number N. 5

(b) Calculate the nth divided difference of l

x
on the nodal points x(, %y, ..., X, . 5

3. (a) Find the inverse of the matrix : 5

32 1]
A=(2 3 2’
12 2|

using LU decomposition method with

Uyy =tgg =tgy =1.

(b) Use secant method to determine the root of

the equation cosx-xe®=0. Take the

initial approximation as xo =0, x, =1 and

perform two iterations of the method. 5

(a) Using the data sin (0.1) = 0.09983 and

sin (0.2) = 0.19867, find an approximate
value of sin (0.15) by Lagrange’s
interpolation. Also obtain a bound on the
truncation error. 5
(b) Use the Euler's method to solve
numerically the initial value problem : 5

¥ =-2xy%, y0)=1
with 2 = 0.2 on the interval [0, 1].

(a ) Using Runge-Kutta fourth order method

with h = 0.1, find an approximate value of
(0.1) for the initial value problem : 5

¥ =xy+yt y0)=1
(b Evaluate the integral :
1.4
I(sinx~logl,x+ex)dx
0.2
using composite trapezoidal rule with
h = 0.2, compare with the exact value. 5
 (a) Given the data :

f(3) = 168,
A7) =120
and f9) =12

If f(k) is estimated as 138 using the

Newton’s form of the interpolating

polynomial, then find the value of k. 5

(b) Solve the system of equations : 5

xX+2y+2=3

3x-2y-4z=-2

2x+3y-z=-6

using Gauss-Elimination method.

(a) Using synthetic division method, check

whether o = 3 is a root of the polynomial

equation : 4

2t rx? -18x%2 —x+12=0

(b) Find the first term of the series whose

second and subsequent terms are

8, 3, 0, — 1, 0 using difference table. 6

(a) Estimate the eigen values of the matrix: 6

1 2 -1]
A=[1 1 1|
1 3 -1

using the Gerschgorin bounds. Also. draw

the rough sketch of the region in which the
eigen values lie.

(b) If: 4

fx)=e™,

show that :
AF(x) = (0 —1ytax
9. (a) From the following data, calculate the

population in the year 1985 : 5

Doar Population
(in’000)

1971 12

1981 15

1991 20
2001 27

2011 49

(b) Using the third order Taylor’'s series

method, find the solution of the initial
value problem : 5

Y=x-y »0)=1
at x=0.1 taking
2~ =0.1.

10. (a) How many terms n be chosen in

Maclaurin’s expansion for e* with an error

less than 1075, —1<x<1 ? 5

(b) Find one root of the equation : 5

x3-2x-5=0
in the interval [2, 3] using Birge-Vieta
method. Perform only one iteration.