Name : …………………………………………….

………………
htt
Roll No. : …………………………………………...……………..
Invigilator’s Signature : ………………………………………..

CS/B.TECH(NEW)(APM/CSE/IT/AUE/CHE/BT/ME
p:/
/PE/CE/CT/LT/TT/FT/SEM-4/M(CS)-401/2012

2012
/q
NUMERICAL METHODS
Time Allotted : 3 Hours Full Marks : 70
pap
The figures in the margin indicate full marks.
Candidates are required to give their answers in their own words
as far as practicable.

GROUP – A
er.

( Multiple Choice Type Questions )

1. Choose the correct alternatives for any ten of the following :
wb

10 × 1 = 10
5
i) If 3
is approximated to 1·6667, then absolute error is

a) 0·000033 b) 0·000043
ut .

c) 0·000034 d) none of these.

1
ii) If f (x ) = then the divided difference f (a , b ) is
x2
a c.

(a + b ) (a − b )
a) b)
(ab )2 (ab )2

1 1 1
c) 2
− 2 d) .
a b a − b2
2
in

4051 [ Turn over

CS/B.TECH(NEW)(APM/CSE/IT/AUE/CHE/BT/ME
/PE/CE/CT/LT/TT/FT/SEM-4/M(CS)-401/2012

Δ2 3
httiii) The value of ( x ) is
E

a) x b) 6x
p:/
c) 3x d) x2 .

dy
iv) If = x + y and y ( 1 ) = 0, then y ( 1. – 1 ) according to
dx
/q
Euler's method is, h = 0 ⋅ 1 ( say ),

a) 0·1 b) 0·3
pap
c) 0·5 d) 0·9.

v) If y 0 = 2 , y1 = 4 , y 2 = 8 , y 4 = 32 then y 3 is equal to

a) 5 b) 15
er.

c) 6 d) 16·5.

vi) The order of h in the error expression of Trapezoidal rule

wb

is

a) 1 b) 2

c) 3 d) 4.
ut .

vii) Regula-Falsi method is

a) conditionally convergent
a c.

b) linearly convergent

c) divergent

d) none of these.
in

4051 2
 CS/B.TECH(NEW)(APM/CSE/IT/AUE/CHE/BT/ME
/PE/CE/CT/LT/TT/FT/SEM-4/M(CS)-401/2012
viii) Pivoting is very much essential because
htt
a) determinant of the coefficient matrix should be

grater than zero

p:/
b) pivot element should not have very large value

compared to the elements of the matrix

/q
c) it reduces the possibility of division by zero
pap
d) change of convergence is higher.

ix) Which of the following is true ?

er.

a) Δn x n = ( n + 1 ) ! b) Δn x n = n !

c) Δn x n = 0 d) Δn x n = n .
wb

x) An n × n matrix A is said to be diagonally dominant if

ut .

n n
a) aii ≥ ∑ aij b) aii ≤ ∑ aij
j =1 j =1
i≠ j i≠ j
a c.

n n
c) aii > ∑ aij d) aii < ∑ aij .
j =1 j =1
i≠j i≠ j
in

4051 3 [ Turn over

CS/B.TECH(NEW)(APM/CSE/IT/AUE/CHE/BT/ME
/PE/CE/CT/LT/TT/FT/SEM-4/M(CS)-401/2012
xi) The condition of convergence of Newton-Raphson's
htt
method is

a) | f ( x ). f ′( x )| < { f ′′( x ) } 2
p:/
b) | f ( x ). f ′′( x )| < { f ′( x ) } 2

c) | f ( x ). f ′( x )| > { f ′′( x ) } 2
/q
d) | f ( x ). f ′′( x )| > { f ′( x ) } 2 .
pap
dy
xii) For = xy and y ( 0 ) = 2 , the value of k 2 according to
dx

Runge-Kutta method of 2nd order is ( h = 0·2 )

er.

a) 0·1 b) 0·01

c) 0·4 d) 0·04.
GROUP – B
wb

( Short Answer Type Questions )

Answer any three of the following. 3 × 5 = 15

2. Given u0 + u6 = 3 , u1 + u5 = 5 , u2 + u 4 = 7 . Find u3 , where v x

ut .

is a function of x.

dy
3. Using the following table find at x = 0 & 1·5.
dx
a c.

x : 0 1 2 3

y : 1 2 11 34
in

4051 4
 CS/B.TECH(NEW)(APM/CSE/IT/AUE/CHE/BT/ME
/PE/CE/CT/LT/TT/FT/SEM-4/M(CS)-401/2012
4. Solve the following system of equations using Gaussian
htt
elimination method :

x +y +z =9
2x − 3y + 4z = 13
p:/
3x + 4y + 5z = 40

1
5. Find the value of ( 19 )3 correct to four decimal points by
/q
Newton-Raphson method.
pap
6. Find the cubic polynomial by Lagrange's interpolation

formula which takes the following value :

x : 0 4 5 8

f(x):
er.
1 2 1 10

GROUP – C
( Long Answer Type Questions )
wb

Answer any three of the following. 3 × 15 = 45

7. a) Find a root of the equation x 4 − x − 10 = 0 that lies

between 1 & 2 using Newton-Raphson method correct to

ut .

3 places of decimal.

b) Solve the system of equations

a c.

x + y + 54z = 110
27x + 6y − z = 85
6x + 15y + 2z = 72

by Gauss-Seidel method. 7+8

in

4051 5 [ Turn over

CS/B.TECH(NEW)(APM/CSE/IT/AUE/CHE/BT/ME
/PE/CE/CT/LT/TT/FT/SEM-4/M(CS)-401/2012
8. a) Solve the following system of equations by
htt
LU-factorization method :

3x + 2y + 7z = 4
p:/
2x + 3y + z = 5
3x + 4y + z = 7

b) Using Runge-Kutta method of order 4, final y ( 0·2 )

/q
dy
given that = 3e x + 2y , y ( 0 ) = 0,taking h = 0·1.
dx
pap
7+8

9. a) Find the root of the equation 3x − cos x − 1 = 0 by

er.
Regula-falsi method, correct to three decimal places.

π
2
b) Evaluate ∫ cos x dx by using (i) Trapezoidal and
wb

1
(ii) Simpson's 3
rd rule, where h = 15°. 7+8
ut .

10. a) Compute y = ( 1 ⋅ 4 ) by Milne's predictor & corrector's

dy 1
method from = (x +y ) where y ( 1 ) = 3 ⋅ 595 ,
dx 2
a c.

y ( 1 ⋅ 1 ) = 3 ⋅ 833 , y ( 1 ⋅ 2 ) = 4 ⋅ 088 , y ( 1 ⋅ 3 ) = 4 ⋅ 362 .

b) Derive Newton's divided difference formula.

in

4051 6
 CS/B.TECH(NEW)(APM/CSE/IT/AUE/CHE/BT/ME
/PE/CE/CT/LT/TT/FT/SEM-4/M(CS)-401/2012
dy
c) Given that = log10 ( x + y ) with the initial condition
htt
dx

that y = 1 when x = 0. Find y for x = 0·2 and x = 0·5

using Euler's modified formula. 5+5+5

p:/
11. a) If y = f (x ) is a polynomial degree 5 with

y 0 = f (0 ) = 0 , y1 = f (1) = 3 , y 2 = f (2 ) = 14 ,
/q
y 3 = f (3 ) = 45 , y 4 = f ( 4 ) = 84 , y 5 = f (5 ) = 170 ,
pap
y 6 = f (6 ) = 258 . It is found that there is one error in

the value of y 3 . Find the correct value of y 3 .

b) Why implicit method is preferred over explicit method

er.

though it requires more computations ?

c) Show that the rate of convergence in Newton-Raphson

wb

method is quadratic. 8+3+4

=============
ut .
a c.
in

4051 7 [ Turn over