EDUCATIVE JEE (MATHEMATICS)

ERRATA FOR SECOND EDITION

Last updated on April 16, 2011

All the errors noticed in the rst edition before the second edition went to press have been corrected in the second edition. Also, the alternate proofs given in the Errata to the rst edition have been incorporated in the second edition. The present errata list the errors that came to light after the second edition went to press. Any errors still remaining may be brought to the attention of the author by e-mail (kdjoshi@math.iitb.ac.in) or by phone (9819961036 or 9713612285). Alternate solutions are also welcome. On p. 176, a shorter, albeit trickier, proof of the identity (15) can be given by recognising the series on the L.H.S. as a telescopic series. Specically, for k 0, rewrite 2n 4k + 1 in the numerator 2nk 2n2k k as (2n 2k + 1) 2k. Then note that the expression is 2nk
n

simply

n n2k 2 k 2n2k 2n2k nk

while the expression

n k1 2n2k+2 nk+1

2nk 2k2n2k k 2nk (2n 2k + n

1)

, upon

simplication, equals

2n2k+2. Hence the k-th term of the

n k 2n2k nk

L.H.S. can be expressed as Ak Ak1 where Ak =

2n2k . Since

A1 = 0, the sum equals Am which is precisely the R.H.S. (Contributed by Gaurav Bhatnagar.) On p. 509, in Exercise (13.19), the condition x should be 2 2 replaced by < x < . However, then there are no values of for 2 2 which f has a (global) maximum and also a (global) minimum. If the problem is taken to mean a local maximum and a local minimum, then 3 3 the answer is ( 2 , 0)(0, 2 ). This is obtained by putting u = sin x, and noting that the derivative of the function g(u) = u3 + u2 must have two distinct roots in the interval (1, 1). (Pointed out by Sameer Kulkarni.) /2 sin((2m 1)u) On p. 668, it is mentioned that the integral du sin u 0 can be evaluated using Chebychev polynomials. Actually, an easy evaluation is possible. Call this integral as Jm for m 1. Then the inte2 cos(2mu) sin u grand of Jm+1 Jm can be written as = 2 cos(2mu). sin u

/2 /2 1 = 0. But then Jm+1 Jm equals 2 cos 2m du = sin(2mu) 0 m 0 Thus we have shown that Jm+1 = Jm for all m 1. A direct calculation gives J1 = /2. This proves (12) on p. 668 and gives an alternate (and a more direct) solution to the problem. The following table lists the minor errors. (A negatively numbered line is to be counted from the bottom.)

Page (line) For 1 1 63 (-18) 2( 2 + 2 ) 183 (2) (1 x + x2 )(1 + x + x2 ) 356 (-10) By2 366 (8) r1 r2 r3 r1 + r2 + r3 368 (-4) is far too 368 (-4) useful. 455 (-19) (at2 , 2ati ) 1 503 (-7) y2 = f (x3 ) 518 (-11) cos A0 518 (-10) cos B0 518 (-9) cos C0 547 (-5) g(x0 ) 564 (-17) g() = f (c) 579 (-15, -9) Comment No. 21 597 (-2) but that g 709 (-17) continuous 941 (-10) system (10) 942 (2) (10) 971 (-20) 16 977 (-2) derivatives 978 (7) 1 + 4 log2 x 979 (2) g(0) and g(1) 986 (-14) (c) 1 990 (10) 1 23 1001 (-8) (d) 0. 1024 (2) 2(a b + b b + c a) 1046 (11) 977 1052 (18) Yaglom, 976

Read i 1 2( 2 + 2 ) (1 x + x2 )n (1 + x + x2 )n BY 2
r1 r2 r3 r1 +r2 +r3

is too useful all by itself. (at2 , 2ati ) i y2 = f (x2 ) sin A0 sin B0 sin C0 f (x0 ) g() = f () Comment No. 19 and that g non-negative and continuous system (11) (11) 1/16 limits 1 + 4 log2 x g(0) and g( 1 ) 2 (c) 2 ln 2 1 1 123 2 (d) 14. 2(a b + b c + c a) 1034 Yaglom, 1034