100% found this document useful (1 vote)
51 views5 pages

UPSC Civil Services Main 1985 - Mathematics Calculus: Sunder Lal

This document contains solutions to questions from past UPSC Civil Services mathematics exams. It summarizes that for a function f(x,y), the partial derivatives fx and fy exist at (0,0) but the function is not continuous there. It also proves relations between beta and gamma functions, and uses Rolle's theorem and Maclaurin's expansion to solve other calculus problems regarding derivatives and integrals. The document provides detailed solutions to exam questions on these calculus topics.

Uploaded by

sayhigaurav07
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
100% found this document useful (1 vote)
51 views5 pages

UPSC Civil Services Main 1985 - Mathematics Calculus: Sunder Lal

This document contains solutions to questions from past UPSC Civil Services mathematics exams. It summarizes that for a function f(x,y), the partial derivatives fx and fy exist at (0,0) but the function is not continuous there. It also proves relations between beta and gamma functions, and uses Rolle's theorem and Maclaurin's expansion to solve other calculus problems regarding derivatives and integrals. The document provides detailed solutions to exam questions on these calculus topics.

Uploaded by

sayhigaurav07
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
You are on page 1/ 5

UPSC Civil Services Main 1985 - Mathematics

Calculus
Sunder Lal
Retired Professor of Mathematics
Panjab University
Chandigarh

January 16, 2010

Question 1(a) If (
xy
x2 +y 2
, (x, y) 6= (0, 0)
f (x, y) =
0, (x, y) = (0, 0)
Show that both the partial derivatives fx , fy exist at (0, 0), but the function is not continuous
there.

Solution. See 2004, question 2(d).

Question 1(b) If for all values of the parameter and some constant n, F (x, y) =
F F
n F (x, y) identically, where F is assumed to be differentiable, prove that x +y =
x y
nF (x, y).

Solution. See 1996, question 2(a).

Question 1(c) Prove the relation between the beta and gamma functions

(m)(n)
(m, n) =
(m + n)

Solution. See 1991, question 2(c).

1
Question 2(a) If a function f defined on [a, b] is continuous on [a, b] and differentiable on
(a, b), and f (a) = f (b), then prove that there exists at least one real number c, a < c < b
such that f 0 (c) = 0.

Solution. This is Rolles theorem. Since f is continuous on [a, b], it is bounded


and attains its maximum and minimum. Since f (a) = f (b) and we assume that f is not a
constant function, it follows that there exists a real number c such that f (c) = M , maximum
of f (without loss of generality) and a < c < b. Then the right hand derivative of f at c is
f (c + h) f (c)
lim+ and clearly it is 0, because f (c + h) f (c) 0 and h > 0. However
h0 h
f (c + h) f (c)
the left hand derivative of f at c is lim 0 because f (c + h) f (c) 0 and
h0 h
0
h < 0. Since f (c) exists, the right hand derivative of f at c equals the left hand derivative
of f at c, which is possible only when both are zero, i.e. f 0 (c) = 0.

Question 2(b) Use Maclaurins expansion to show that

x2 x3 x4
log(1 + x) = x + + ...
2 3 4
Hence find the value of log(1 + x + x2 + x3 + x4 ).

Solution. Since log(1 + x) possesses continuous derivatives of all orders for every value of
x for which 1 + x > 0 i.e. x > 1, we have

x2 00 xn1 (n1)
f (x) = log(1 + x) = f (0) + xf 0 (0) + f (0) + . . . + f (0) + Rn
2! (n 1)!

where
xn (n) xn (n 1)!
Rn = f (x) = (1)n1 , 0<<1
n! n! (1 + x)n
(1)n1 (n 1)!
as f (n) (x) for f (x) = log(1 + x) is .
(x + 1)n
x 1
Let 0 < x 1, so that 0 < < 1, then |Rn | < n
and therefore Rn 0 as n .
1 + x
Let 1 < x 0, we consider the expansion

x2 00 xn1 (n1)
f (x) = log(1 + x) = f (0) + xf 0 (0) + f (0) + . . . + f (0) + Rn
2! (n 1)!

where
n1
xn

n1 (n) n n 1 1
Rn = (1 ) f (x) = (1) x , 0<<1
(n 1)! 1 + x 1 + x

2
this is Cauchys form of the remainder note that we consider Cauchys form of the
remainder as Lagranges form of the remainder does not enable us to prove that Rn 0 as
n . when 1 < x < 0.
1 1 1
Since 0 < < 1 and |x| < 1, 0 < < 1. Also, < . Now since xn 0
1 + x 1 + x 1 |x|
as n , we get Rn 0 as n .
Now using the fact that f (n) (0) = (1)n1 (n 1)!, we get the required result
x2 x3 x4
log(1 + x) = x + + ...
2 3 4

1 x5
log(1 + x + x2 + x3 + x4 ) = log( ) = log(1 x5 ) log(1 x)
1x
x10 x15 x2 x3 x4
= (x5 + + + . . .) + (x + + + + . . .)
2 3 2 3 4
x2 x3 x4 1  x6 x7 x8 x9 x10  1 
= x+ + + + x5 1 + + + + + 1 + ...
2 3 4 5 6 7 8 9 2 5
which is the required expansion.

Question 2(c) Find the volume generated by revolving y 2 = 4ax about the latus rectum.

Solution.

Let P be the point (x, y), M be the foot of


the perpendicular on the latus rectum x = a. P M
Then P M = a x. SM = y, where S is
the focus (a, 0). Because of symmetry we S (a, 0)
can confine ourselves to the first quadrant, x=a
0 y 2a.
y 2 = 4ax

Z Z 2a
2
V = 2 (P M ) d(SM ) = 2 (a2 2ax + x2 ) dy
0
2a 
y2 y4 
Z
= 2 + a2 dy
0 2 16a2
2a
y3 y5 8a3 32a5
  
2 3
= 2 a y + = 2 2a +
6 80a2 0 6 80a2
30 20 + 6 32a3
= 2a3 =
15 15

3
Paper II
Question 3(a) Show that for the function
( 2
x y2
x2 +y 2
, (x, y) 6= (0, 0)
f (x, y) =
0, x = 0, y = 0
1. fx is not differentable at (0, 0).
2. fyx is not continuous at (0, 0).
3. fxy (0, 0) = fyx (0, 0).
Solution.
1. For (x, y) 6= (0, 0),
(x2 + y 2 )2xy 2 x2 y 2 (2x) 2xy 4
fx (x, y) = = 2
(x2 + y 2 )2 (x + y 2 )2
2.
h2 y 2
f (h, y) f (0, y) h2 +y 2
0 hy 2
fx (0, y) = lim = lim = lim =0
h0 h h0 h h0 h2 + y 2

Thus fx (0, y) = 0 for y 6= 0.


3.
f (x + h, 0) f (x, 0)
fx (x, 0) = lim =0
h0 h
In particular, fx (0, 0) = 0.
fx (0,k)fx (0,0)
4. fyx (0, 0) = limk0 k
=0
f (x,k)f (x,0) x2 k
5. fy (x, 0) = limk0 k
= limk0 x2 +k2
= 0 when x 6= 0.
f (0,k)f (0,0)
6. fy (0, 0) = limk0 k
= 0.
fy (h,0)fy (0,0)
7. fxy (0, 0) = limh0 h
= 0.
Thus we have shown that fxy (0, 0) = fyx (0, 0) = 0.
8. For (x, y) 6= (0, 0),
2xy 4
fyx (x, y) = fx (x, y) =
y y (x2 + y 2 )2
8xy 3 (x2 + y 2 )2 2xy 4 2(x2 + y 2 ) 2y
=
(x2 + y 2 )4
3
8xy 3 (x2 + y 2 ) 8xy 5 8x3 y 3

xy
= = 2 =8 2
(x2 + y 2 )3 (x + y 2 )3 x + y2
Now from question 1(a) above, it follows that fyx is not continuous at (0, 0).

4
9. If fx were differentiable at (0, 0), we would have

fx (h, k) = fx (0, 0) + hfxx (0, 0) + kfyx (0, 0) + h2 + k 2 (h, k)

where limh0,k0 (h, k) = 0. Substituting the known values, we get

2hk 4 1
(h, k) =
(h2 + k 2 )2 h2 + k 2
2m4
Now we can see that limh0,k0 (h, k) does not exist if k = mh, (h, mh) = 5 ,
(1+m2 ) 2
so limh0 (h, mh) is different for different values of m (say 0, 1).

Thus we have shown all the three results required.

You might also like