lfH1lll!

:::i

Duration: Three hours MaximumMarks: 150

)- Read the following instructions carefully.

SECTION A (75 Marks)
1. All answers must be written in English.
2. This question paper contains Two Sections: 'A'
and 'B'. 1. - MA. This question consist of twenty five
3. Section A consists of two questions of multiple sub-questions (1.1 - 1.25) of one mark
choice type. Question 1 consists of Twenty Five each. For each of these sub-questions, four
sub-questions of One mark each and Question 2
possible answers (A,B,C and D) are given,
consists of Twenty Five sub-questions of Two
marks each. out of which only one is correct. Answer
4. Answer Section A only on the special machine- each sub-question by darkening the
gradable Objective Response Sheet (ORS). appropriate bubble on the objective
Questions in Section A will not be graded if response sheet (ORS) using an HB pencil.
answered elsewhere. Do not use the ORS for any rough work.
5. Write your name, registration number and the name Use the Answer book, if you need to do
of the centre at the specified locations on the right
half of the ORS for Section A
any rough work.
6. Using a HB pencil, darken the appropriate bubble 1.1 The dimension of the vector space of all
under each digit of your registration number. 3x3 real symmetric matrices is
7. Questions in Section A are to be answered by (a) 3
darkening the appropriate bubble (marked A, B, C
or D) using a HB pencil against the question (b) 9
number on the left hand side of the ORS. In case, (c) 6
you wish to change an answer, erase the old answer
completely using a good soft eraser. (d) 4
8. The ORS will be collected after 120 minutes, you 1.2 Let A be a non-zero upper triangular
may start answering section B. matrix all of whose eigen values are 0.
9. There will be Negative marking in Section A For Then I+ A is
each wrong answer to 1 and 2 mark sub-questions,
0.25 and 0.5 marks will be deduced respectively. (a) Invertible
More than one answer marked against a question (b) Singular
will be deemed as an incorrect response and will be
negatively marked. (c) Idempotent
10. Answer questions in Section Bin the answer book. (d) Nilpotent
Section B consist of Twenty questions of Five
1.3 The eigen values of a skew-symmetric
marks each. Any Fifteen out of them have to
answered. If more number of questions are matrix are
attempted, score off the answers not to be (a) Negative
evaluated, else only the first fifteen un scored
answers will be considered.
(b) Real
11. Answer for each question in Section B should be (c) Of a absolute value 1
started on a fresh page. Question numbers must be (d) Purely imaginary or zero
written legibly and correctly in the answer book.
12. In all 5 marks questions (Sections B), clearly show 1.4 The function f (z) = z 2 maps the first
the important steps in your answers. These quadrant onto
intermediate steps will carry partial credit.
)- Note: The Symbols Z, R, and C denote the set of (a) Itself
all integers, real numbers and complex numbers (b) Upper half plane
respectively. Vector quantities are denoted by bold
letters.
(c) Third quadrant
(d) Right half plane
1.5 The radius of convergence of the power outward unit normal to S. Then the

series of the function

1
f (z) = - - about integral ILr .n dS equals
1-z
1 . (a) V
Z =-IS
4 (b) v
(a) 1 3
(c) 3V
(b) I_ (d) 0
4
1.10 Let G be a cyclic group of order 6. Then
3
(c)- the number of elements g E G such that
4
G =< g >is
(d) 0
(a) 5
1.6 Let T be any circle enclosing the origin
and oriented counter-clockwise. (b) 3

.
Then the value of the mtegra1 -cosz
2
-
dz IS
. I (c) 4
(d) 2
r z
1.11 The number of elements of order 5 in the
(a) 2;ri symmetric group Ss is
(b) 0 (a) 5
(c)- 2;ri (b) 20
(d) Undefined (c) 24
1.7 Suppose S1 , S 2 and S3 are measurable (d) 12
subsets of [ 0, 1], each of measure %, such 1.12 The order of the element (2, 2) m
that the measure of sl u s2 u s3 is 1. Then, Z 4 xZ6 is
the measure of sl ! l s2 ! l s3 lies in (a) 2

(a) ｛ｯＬ Ｑ ｾ｝＠

(b) 6
(c) 4
(d) 12
(b) ｛ＱｾＧｩ｝＠ 1.13 Which of the following Banach spaces is
not separable?
(c)[i,±] (a) L1 [0,1]
(b) r) [o,1]
(d) [±,1] 2
(c) L [0,1]
1. 8 Let f :[ 0, 1] ----+ R be a bounded Riemann (d) C(0,1]
integrable function and let g : R ----+ R be
1.14 For a subset A of a metric space, which of
continuous. Then go f is the following implies the other three?
(a) Riemann integrable (a) A is closed
(b) Continuous (b) A is bounded
(c) Lebesgue integrable, but not Riemann (c) Closure of B IS compact for every
integrable Be A
(d) Not necessarily measurable (d) A is compact
1.9 Let V be the volume of a region bounded 1.15 Let n be a nonnegative integer. The eigen
by a smooth closed surfaceS. Let r denote values of the Sturm-Lioubille problem?
the position vector and ii denote the
 d2y e-x 2 /4t
dx2 +Ay=O, (d) Ji
With boundary conditions 1.20 Which of the following is elliptic?
y(o) = y(2Jr), dy (o) = dy (2Jr) are (a) Laplace equation
dx dx (b) Wave equation
(a) n
(c) Heat equation
(b) n2JZ"2
(d) uxx+2uxy-4uyy=O
(c) 7t2
2 1.21 Consider the motion of a three-particle
(d) n system in the 3-dimensional space.
1.16 The Bessel's function ｻｊ Ｐ Ｈ｡ｫｸＩｽ］ｾ＠ with Suppose that all pairs of particles are at
invariant distance apart, so that the system
ak denoting the k-th zero of J 0 ( x) form constitutes a rigid body. For such a system,
an orthogonal system on [ 0, 1] with respect the number of degrees of freedom is
to weight function (a) 6
(a) 1 (b) 3
(c) 1
(b) x 2
(d) 9
(c) x
1.22 If f (x) has an isolated zero of
(d) Fx
multiplicity 3 at x = q, and the iteration
1.17 Linear combinations of solutions of an
ordinary differential equation are also 3J( xn)
solutions if the differential equation is xn+l = xn - fl ( xn) ' n = 0,1, 2, ....
(a) Linear non homogeneous
Converges to q, then the rate of
(b) Linear homogeneous
convergence IS
(c) Nonlinear homogeneous
(a) Linear
(d) Nonlinear non homogeneous
(b) Faster than linear but slower than
1.18 Which of the following, concerning the quadratic
solution of the Neumann problem for
(c) Quadratic
Laplace's equation, on a smooth bounded
domain, is true? (d) Cubic
(a) Solution is unique 1.23 The best possible error estimate in the
Gauss-Hermite formula with 3 points, for
(b) Solution is unique up to an additive
constant calculating the integral L: x 4 e-x' dx is
(c) Solution IS unique up to a (a) 0
multiplicative constant
(b) 0.30
(d) No conclusion can be drawn about
uniqueness (c) 0.65
1.19 Which of the following satisfies the heat (d) 1.20
equation (without source terms and with A
1.24 Let P (X = n) = 2 ( )' where A is an
diffusion constant 1) m one space n n+1
dimension? appropriate constant. Then E(X) is

(a) ｳｩｮＨＺｾＩ＠ (a) 2A+1

(b) A
(b) e sin x 1 (c) oo
(c) x 2 -t (d) 2A
1.25 Let x be a non-optimal feasible solution of
2.4 Forthefunction f(z)=sin..!.,z=O is a
a linear programmmg maxnmzation z
problem and y a dual feasible solution.,
(a) Removable Singularity
Then
(b) Simple pole
(a) The primal objective value at x is
greater than the dual objective value at y (c) Branch point
(b) The primal objective value at x could (d) Essential singularity
equal the dual objective value at y 2.5 Pick out the largest of the sets given below
(c) The primal objective value at xis less on which the sequence of functions
than the dual objective value at y
(d) The dual could be unbounded
{e-ncos'x ｛ｾ
＠Ｑ converges uniformly
(a) [0, ｾＩｵＨ
Ｑ
2. -MA. This question consists of Twenty ［ｾ＠ ,;r]
Five sub-questions (2.1 - 2.25) of Two
marks each. For each of these sub- Jr Jr
(b) [0,-)u(-,;r]
question, four possible answers (A,B,C 2 2
and D) are given, out of which only one is
Jr Jr Jr
correct. Answer each sub-question by (c) [0,--S)u(-+S,;r], 0<6 < -
darkening the appropriate bubble on the 2 2 100
Objective Response Sheet (ORS) using a (d) [O,;r]
soft HB pencil. Do not use the ORS for
Let f : R ----+ R be a smooth function with
2
2.6
any rough work. Use the Answer book, if
positive definite Hessian at every point.
you need to do any rough work.
2.1 Let A be a 3x3 matrix with eigenvalues 1, Let ( a,b) E R 2 be a critical point of f.
100
-1, 0. Then the determinant of I+ A is Then
(a) 6 (a) f has a global minimum at (a, b)
(b) 4 (b) f has a local, but not a global
(c) 9 minimum at (a, b)
(d) 100 (c) fhas a local, but not a global
2.2 Let A be a 2x2 orthogonal matrix of trace maximum at (a, b)
and determinant 1. Then the angle (d) fhas a global maximum at (a, b)
between Au and u ( u = [1 0 r) is 2.7 Let G be a group of order 30. Let A and B
be normal subgroups of orders 2 and 5
(a) 15° respectively. Then the order of the group
(b) 30° G/AB is
(c) 45° (a) 10
(d) 60° (b) 3
Let w ( z) = az + b and f (z) = a z + fJ be (c) 2
2.3
cz+d yz+6 (d) 5
bilinear (Mobius) transformations. Then 2.8 Let m and n be coprime natural numbers.
the following IS also a bilinear Then the kernel of the ring homomorphism
transformation ¢:Z ---+Zmxzn, defined by ¢(x)=(x,x)
(a) f(z)w(z) IS

(b) f(w(z)) (a) mZ

(b) mnZ
(c) f(z)+w(z)
(c) nZ
1 (d) z
(d) f (z) + w ( z)
2.9 Consider the Banach space C [ 0, 1r] with derivative of u on the boundary of the unit
disc)
the supremum norm The norm of the
(a) 2;r
linear functional l: C [ 0, 1r]----+ R given by

l(f) = r 2
f(x)sin xdx is
(b) 1
(c) ;r
(a) 1 (d) 0
2.15 Let u be a solution of the initial value
Ｈ｢Ｉｾ＠ problem
2
(c) n a2u a 2u 2 au
- - - = 0; u(x,O) = x ,-(x,O) = 0.
2 2
(d) 2;r at ax at
2.10 The topology on the real line R generated Then u ( 0,1) equals
by left-open right-closed intervals (a, b] is
(a) 1
(a) Strictly coarser than the usual topology
(b) 0
(b) Strictly finer than the usual topology
(c) 2
(c) Not comparable with the usual
topology (d) I_
2
(d) Same as the usual topology
2.16 Euler's equation of motion of a rigid body
2.11 Let X, Y be topological spaces and
about a fixed point 0 in the absence of
f: X----+ Y be a continuous and bijective
external forces are
map.
Then f is a homeomorphism, if A dw1 -(B-C)w w = 0
dt 2 3 '

(a) X andY are compact

(b) X is Hausdorff and Y is compact Bdw 2 -(C-A)ww =0
dt 3 I '

(c) X is compact and Y is Hausdorff

dw 3 (
C--A-B )
ww =0
(d) X andY are Hausdorff dt I 2 '

2.12 If the integrating factor of Where w1 , w 2 , w3 are components of the

(X y7 2
+ 3 y) dx + ( 3x8 y - X) dy = 0 is Xm yn , angular velocity of the rigid body in the
then direction of principal axes of inertia at 0;
(a) m=-7, n=1 A, B, C are principal moments of inertia at
0. LetT denote the kinetic energy during
(b) m = 1, n = -7 the motion and h the magnitude of the
(c) m=n=O angular momentum of the body 0. Then,
during the motion
(d) m =n = 1
(a) T and h both vary
2.13 The initial value problem
(b) T varies but h remains constant
(x -x): =(2x-1)y,y(x )=y
2
0 0 (c) T remains constant but h varies
(d) T and h both remain constant
Has a unique solution if (x 0 ,y0 ) equals
2.17 The fourth divided difference of the
(a) (2,1) polynomial 3x 3 + 11x 2 + Sx + 11 over the
(b) (1,1) points x = 0, 1, 4, 6 and 7 is
(c) (0,0) (a) 18
(d) (b) 11
2 2
2.14 If u harmonic on {( x,y) I x + y ::::; 1}, then (c) 3
2J!" au au (d) 0
f
0
-de equals (where - is the normal
an an
2.18 The polynomial of least degree (a) ac < O;b,d E (O,oo)
interpolating the data
( 0, 4), ( 1, 5), ( 2, 8), ( 3, 13) is (b) ac<O;b,dE(-oo,O)

(a) 4 (c) ac < O;b,d E R

(b) 3 (d) ac < O;b,d E R
(c) 2 2.23 There are two identical locks, with two
(d) 1 identical keys, and the keys are among the
six different ones which a person carries in
2.19 For the matrix
h is pocket. In a hurry he drops one key
0 1 -1 1 somewhere. Then the probability that the
1 0 0 2 locks can still be opened by drawing one
-1 0 0 0 key at random is equal to

1 2 0 0 1
(a) -
3
The bound for the eigen values predicted
by Gershgorin' s theorem is (b) 2
(a) 3 6
(b) 1 1
(c) -
(c) 2 30
(d) 4 (d)_!_
2.20 An external of the functional 12
2.24 The system Ax::::; 0, where A is an n x n

I[y(x)J = J: F( x,y,: )dx;y(a) = y1 ,y(b) = Y2

matrix,
(a) May not have a nonzero solution
satisfies Euler's equation, which in general (b) Always has a nonzero solution
(a) Is a second order linear ordinary (c) Always has at least 2 linearly
differential equation (ODE) independent solutions
(b) Is a nonlinear ODE of order greater (d) Always has at least n linearly
than two independent solutions
(c) Admits a unique solution satisfying the 2.25 Consider the following linear program
conditions y( a)= y1 ,y(b) = Y2 n
P: max 2:C1x1 subject to
(d) May not admit a solution satisfying the ｪｾｬ＠

conditions y( a)= yPy(b) = Y2

2.21 A lot of 1000 screws contains 1% with
major defects and 5% with minor defects. Suppose that we are keeping the ｣ｾｳ＠ and
If 50 screws are picked at random and
inspected, then the ordered pair (expected ｡ｾｳ＠ fixed and varying the b:S . Suppose
number of major defectives, expected that P is unbounded for some set of the
number of minor defectives) is parameter values b; . Then, for every
(a) (1,5) choice of b;s ,
(b) (2.5, 0.5)
(a) P is unbounded or infeasible
(c) (0.5, 2.5)
(b) P is unbounded
(d) (5,1)
(c) The dual problem to P has a finite
2.22 The sample correlation of the transformed optimum
random variables aX+ b and cY +dis
(d) The dual problem to P is unbounded
same as that of X andY provided
 SECTION B (75 Marks) 10. Find the general solution of the differential
equation
This Section consists of Twenty questions of Five
marks each. Any Fifteen out of them have to be d4y .
---y=xsmx.
4
answered on the Answer book provided. dx
(15x5 = 75) 11. Find the general solution of the differential
equation
3. Let Jn be the n x n matrix each of whose 2 d 2y dy
(X -1) dx 2 + (X -1) dx - 4 y = 0
entries equals 1. Find the nullity and the
characteristic polynomial of Jn. In powers of (x-1) using the Frobenius
4. Evaluate the integral method.
d Solve the initial value problem
I00

2
x 2 ,a> 0,
2
12.
au au
2 2
-oo ( x + a )
- 2- - =
2
0 t >0 XE (-oo oo)
at ax ' ' '
By the method of residue calculus.
au X
5. Let a, b be real numbers with 0 < a< b. u(x,O) = x, -(x,O) = - -2
at 1+x
Define sequences {an} and {bn}
13. Consider the initial value problem
recursively by
au au
-+u-=0, y>O,xER,
an+l ｾ｡ｮ｢＠ and where ay ax

a 1 =a, b1 =b U(X, 0) = j (X)

Show that {an} is an increasing sequence, Show that the solution is constant along
the characteristics. Hence deduce that if f
{bn} is a decreasing sequence, and both is decreasing monotonically, the solution
converge to the same limit. cannot exist as a single valued function for
6. Let f (t) be a real-valued continuous ally> 0.
14. Consider a pendulum consisting of a bob
function on [0,1] such that
of mass m at the end of a rod of 1ength a. If
I

I f(t)tndt = 0, for all n = 0,1,2, .. the bob is pulled to one side through an
angle a to the downward vertical and
0
released, show that the time required for
Prove that f (t) vanishes identically. one complete oscillation is given by
JC
7. Let p be a prime and q be a prime divisor
of 2r -1. Find the order of 2 (the residue T- 4 Jai2 d¢
class of 2) in the multiplicative group G of - fi 0 ｾＮｨ＠ -k 2
sin 2 e
non zero residue classes of integers
modulo q. Conclude that q > p. Where k ］ｳｩｮＨｾ＠ J and g IS the
8. Let X be an infinite dimensional Banach acceleration due to gravity.
space. Prove that X can not have countable
Assume that the mass of the rod and the air
dimension as a vector space.
resistance are negligible.
9. Show that X IS Hausdorff topological
15. Determine the LU decomposition of the
space if and only if the diagonal 11 defined matrix
by
j
l
11 = { (X, y) EX X X I X= y} 5 -2 -3
A= 20 -5 -13
Is a closed subset of X x X (with product
35 -5 -17
topology).
 With L having all its diagonal entries 1; Subject to
and hence solve the system 4

Ia;x; :::=;ao,
AX=[O 2 13r.
16. Using the Runge-Kutta method of order 4
and taking the step size h = 0.1, determine
y(0.1), where y(x) is the solution of Where a; > 0, X; > 0 for i = 1, 2, 3, 4 and
a0 > 0
dy +2xy 2 =0, y(0)=1
dx (i) Write the dual of this Linear
17. Solve the integral equation programming Problem. (3)
I (ii) Assuming
J
¢(x) = x+ (x-¢')¢(¢')d¢'.
.s_>s_>s_>.s_
- - - '
a1 a2 a3 a4
18. Let X 1 ,X2 , .... ,X100 be independent and
Show that the feasible solution
identical Poisson random variables with
100 a -a -a
parameter A= 0.03. Let S = Ix;. Use .x; = x 2 = 1, x3 = 0 1 2 , x 4 = 0,
a3
the Central Limit Theorem to evaluate Is an optimal solution (2)
P( {S 23}) and compare the result with 22. Consider the optimal assignment problem,
the exact probability of the event {S 2 3}. in which n persons R,,P2 .......... ,Pn are to
be assigned n jobs J 1 ,J2 , ...... ,Jn and
19. Let X 1 ,X2 , ...... ,Xn be a random sample
where the effectiveness rating of the
from exponential density fe (X) = ()e-ex, person Pi for the job Ij is aif > 0. The
x 2 0, () > () . Find the maximum likelihood objective is to find an assignment of
estimate (MLE) of 8. Also, find the MLE persons to jobs, that is, a permutation
of P8 ( X 1 2 1) . Further show that both the a-: {1,2, ...... ,n}---+ {1,2, ....... ,n} which
estimators are consistent. assigns person Pi to job Ju(i)' so as to
20. (a) Let ｘｾ＠ ｂＨｾＬｰ Ｑ Ｉ＠ and Y ｾ＠ B(n2 ,p2 ) n

where P1 and P2 are unknowns. Further, let

maximize the total effectiveness I
ｩｾｬ＠
aiu(i) .

X and Y be statistically independent. Show that in any optimal assignment, at

Construct an approximate 100 (1 -a) % least one person is assigned a job at which
confidence interval for (p1 - pJ . (3) he is best.

(b) An antibiotic for pneumonia was

injected into 100 patients with kidney
malfunctions (uremic patients) and into
100 patients with no kidney malfunctions
(normal patients). Some allergic reaction
developed in 38 of the uremic patients and
in 21 of the normal patients. Use the result
in part (a) to construct a 95% confidence
interval for the difference between the two
population proportions (Use the
appropriate table value
Z 0025 = 1.960, Z 005 = 1.645). (2)
21. Consider the Linear program
4

maxic;x;,
ｩｾｬ＠