MADHAVA MATHEMATICS COMPETITION, January 8, 2017

Solutions and scheme of marking

N.B.: Part I carries 20 marks, Part II carries 30 marks and Part III carries 50
marks.
Part I
N.B. Each question in Part I carries 2 marks.
√
1. The number 2eiπ is:
A) a rational number.
B) an irrational number.
C) a purely imaginary number.
D) a complex number of the type a + ib where a, b are non-zero real numbers.
Answer: B √
The number is − 2 using the relation eiπ = −1.
 
0 0 0 1
 1 0 0 0  4
2. Let P =  0 1 0 0 . The rank of P is: A) 1 B) 2 C) 3 D) 4.


0 0 1 0
Answer: D
Det P 6= 0, ⇒ Det P 4 6= 0. Thus rank of P 4 is 4.
dy
3. Let y1 (x) and y2 (x) be the solutions of the differentiable equation = y + 17 with
dx
initial conditions y1 (0) = 0, y2 (0) = 1. Which of the following statements is true?
A) y1 and y2 will never intersect.
B) y1 and y2 will intersect at x = e.
C) y1 and y2 will intersect at x = 17.
D) y1 and y2 will intersect at x = 1.
Answer: A
Solving the differentiable equation we get y1 = −17 + 17ex and y2 = −17 + 18ex . The
two curves never intersect.

4. Suppose f and g are differentiable functions and h(x) = f (x)g(x). Let h(1) = 24,
g(1) = 6, f 0 (1) = −2, h0 (1) = 20. Then the value of g 0 (1) is
A) 8 B) 4 C) 2 D) 16.
Answer: A
0 0 0
h(x) = f (x)g(x). Thus we get h (x) = f (x)g(x) + f (x)g (x)
0 0 0 0
h (1) = f (1)g(1) + f (1)g (1). Therefore 20 = (−2)(6) + f (1)g (1).
0
f (1)g (1) = 32. Now h(1) = f (1)g(1). Therefore 24 = 6f (1).
0
Thus f (1) = 4 and g (1) = 8.

5. In how many regions is the plane divided when the following equations are graphed,
not considering the axes? y = x2 , y = 2x
A) 3 B) 4 C) 5 D) 6.
Answer: D
Plot graph of the two functions y = x2 and y = 2x .

6. For 0 ≤ x < 2π, the number of solutions of the equation sin2 x+3 sin x cos x+2 cos2 x = 0
is
A) 1 B) 2 C) 3 D) 4.
Answer: D
Note cos2 (x) 6= 0. Dividing by cos2 (x) we get tan2 (x)+3 tan x+2 = 0. Thus tan x = −1
or −2. Since tan x has period π and range of tan x is (−∞, ∞), the number of solutions
of the given equation in the interval 0 ≤ x < 2π is equal to 4.

1
 7. The minimum value of the function f (x) = xx , x ∈ (0, ∞) is
 1  1
1 10 1 1 1 e
A) B) 10 10 C) D) .
10 e e
Answer: D
y = xx . Taking log on both sides ln y = x ln x.
1 dy dy
= ln x + 1. Solving = 0 ⇒ lne x = −1. Thus minimum value is obtained at
y dx dx
1
d2 y 1
 
1 e
x = 1/e and is . Note that ( ) > 0.
e dx2 e
00
8. Let f be a twice differentiable function on R. Also f (x) > 0 for all x ∈ R. Which of
the following statements is true?
A) f (x) = 0 has exactly two solutions on R.
0
B) f (x) = 0 has a positive solution if f (0) = 0 and f (0) = 0.
0
C) f (x) = 0 has no positive solution if f (0) = 0 and f (0) > 0.
0
D) f (x) = 0 has no positive solution if f (0) = 0 and f (0) < 0.
Answer: C
00 0 0 0
f (x) > 0 ⇒ f (x) is increasing. Also f (0) > 0 ⇒ f (x) > 0 if x > 0. ⇒ f (x) = 0 has
no positive solution.

1 2 1 2 1 2
     
2 2 27
9. If x + x + 1 = 0, then the value of x + + x + 2 + . . . + x + 27 is
x x x
A) 27 B) 54 C) 0 D) -27.
Answer:  B
1 2 1 2 1 2
   
2 27
x+ + x + 2 + · · · + x + 27
x x x
 2 2  4 2  54 2
x +1 x +1 x +1
= + + ··· +
x x2 x27
2 2
= 9((−1) + (−1) + 2 ) = 54.2

 
−2 −1
10. Let M= . Then M 2017 =
3 1
−22017 −1
       
1 0 −2 3 −2 −1
A) B) C) D) .
0 1 32017 1 −1 1 3 1
Answer: D
Note M 3 = I . Thus M 2016 = I and M 2017 = M.

Part II
N.B. Each question in Part II carries 6 marks.
1 1 1
1. Let a, b, c be real numbers such that a + b + c = + + and abc = 1. Prove that at
a b c
least one of a, b, c is 1.
bc + ac + ab
Solution: Let λ = a + b + c. Then λ = a + b + c = = bc + ac + ab. 2
abc
marks
The numbers a, b, c are roots of the polynomial x3 − λx2 + λx − 1. Observe that x = 1
is one of the roots of this polynomial. 4 marks

2. Let c1 , c2 , . . . , c9 be the zeros of the polynomial z 9 − 6z 7 + 12z 6 + 18z 4 − 24z 3 + 30z 2 −

X9
z + 2017. If S(z) = |z − ck |2 , then prove that S(z) is constant on the circle |z| = 100.
k=1
9
X
Solution: Observe that ck = 0. 2 marks
k=1

2
 9
X 9
X 9
X 9
X
2
S(z) = |z − ck | = (z − ck )(z − ck ) = (zz − zck − ck z + ck ck ) = |z|2 +
k=1 k=1 k=1 k=1
9
X 9
X 9
X 9
X 9
X 9
X 9
X
|ck |2 − z ck − z ck = |z|2 + |ck |2 = (100)2 + |ck |2 = constant. 4
k=1 k=1 k=1 k=1 k=1 k=1 k=1
marks
 
1 00
3. Let f be a monic polynomial with real coefficients. Let lim f (x) = lim f and
x→∞ x→∞ x
f (x) ≥ f (1) for all x ∈ R. Find f.
Solution: Let f (x) = a0 + a1 x + a2 x2 + · · · + an xn . Then f 0 (x) = a1 + 2a2 x +
1
3a3 x2 + · · · + nan xn−1 and f 00 (x) = 2a2 + 6a3 x + · · · + n(n − 1)an xn−2 . Also f ( ) =
  x
1 1 n 1 00
a0 + a1 ( ) + · · · + an ( ) . Now lim f = a0 . Therefore lim f (x) = a0 . Hence
x x x→∞ x x→∞
f 00 (x) = 2a2 and a3 = a4 = · · · = an = 0.   3 marks
00 1
Since f is monic, a2 = 1. Now lim f (x) = lim f implies a0 = 2a2 = 2. Also
x→∞ x→∞ x
f (x) ≥ f (1) for all x ∈ R implies f has minimum at x = 1. Therefore f 0 (1) = 0. Hence
a1 + 2a2 = 0. Therefore a1 = −2. Hence f (x) = x2 − 2x + 2. 3 marks

4. Call a set of integers non − isolated if for every a ∈ A at least one of the numbers a − 1
and a + 1 also belongs to A. Prove that the number of 5−element non − isolated subsets
of {1, 2, . . . , n} is (n − 4)2 .
Solution: Let {a1 , a2 , a3 , a4 , a5 } be 5-element non − isolated subset of {1, 2, . . . , n}
such that a1 < a2 < a3 < a4 < a5 . Then a2 = a1 + 1, a4 = a5 − 1. Further a3 = a2 + 1
or a3 = a4 − 1. Clearly 1 ≤ a1 ≤ n − 4. For each choice of a1 , a5 has (n − 3) − a1 choices.
For a3 , there are 2 choices. So total number of such sets is 2[(n − 4) + (n − 5) + · · · + 1] =
(n − 4)(n − 3)
2 = (n − 4)(n − 3). 4 marks
2
But a3 = a2 + 1 as well as a3 = a4 − 1 gets counted twice. So total number of such sets
is (n − 4)(n − 3) − (n − 4) = (n − 4)(n − 4) = (n − 4)2 . 2

5. Find all positive integers n for which a permutation a1 , a2 , . . . , an of {1, 2, . . . , n} can be

found such that 0, a1 , a1 +a2 , a1 +a2 +a3 , . . . , a1 +a2 +. . .+an leave distinct remainders
modulo n + 1.
Solution: Let a1 , a2 , . . . , an be a permutation of {1, 2, . . . , n} such that 0, a1 , a1 +
a2 , a1 + a2 + a3 , . . . , a1 + a2 + . . . + an leave distinct remainders modulo n + 1. If n is
n(n + 1)
even, then a1 + a2 + · · · + an = ≡ 0( mod n + 1). So remainders modulo n + 1
2
can not be distinct. 3 marks
Let n be odd. Then choose a1 = 1, a2 = n − 1, a3 = 3, a4 = n − 3 and so on. Then
a1 ≡ 1( mod n + 1), a1 + a2 ≡ n( mod n + 1), a1 + a2 + a3 = n + 3 ≡ 2( mod n +
1), a1 + a2 + a3 + a4 = 2n ≡ n − 1( mod n + 1), · · · . This gives required permutation.
3 marks

Part III

1. Do there exist 100 lines in the plane, no three concurrent such that they intersect ex-
actly in 2017 points? [12]
Solution: Consider k sets of parallel lines having respectively m1 , m2 , · · · , mk lines
Xeach set. Then we need to solve the equations m1 + m2 + · · · + mk = 100 and
in
mi mj = 2017. 4 marks
i<j
X 2 X X
mi = m2i + 2 mi mj
i<j

3
 P 2
Therefore
P P −2 2(2017) = 5966. Thus we need to find 100 numbers satis-
mi = 10, 000
fying mi = 100 and mi = 5966. 3 marks
We can get any one of the following solutions:

(a) 75, 18, 3, 2, 2

(b) 77, 3, 2, 2, 2, 2, 1(12 times)
(c) 77, 4, 2, 1(17 times) 5 marks

2. On the parabola y = x2 , a sequence of points Pn (xn , yn ) is selected recursively where

the points P1 , P2 are arbitrarily selected distinct points. Having selected Pn , tangents
drawn at Pn−1 and Pn meet at say Qn . Suppose Pn+1 is the point of intersection of
y = x2 and the line passing through Qn parallel to Y-axis. Under what conditions on
P1 , P2

(a) both the sequences {xn } and {yn } converge?

(b) {xn } and {yn } both converge to 0? [13]

Solution:

(a) Since yn = x2n , it is enough to discuss the convergence of {xn }. Tangents at xn , xn−1
are given by y = x2n + 2xn (x − xn ) = 2xn x − x2n andy = 2xn−1 x − x2n−1 . 
xn−1 + xn xn−1 + xn xn−1 + xn 2
Solving we get, xn+1 = . Therefore Pn+1 ,( ) .
2 2 2
4 marks
Now all xn are within the interval [x1 , x2 ] and all are distinct. Hence {xn } con-
verges. 3
marks
(b) We first consider a special case where x1 = 0, x2 = 1. Then the sequence {xn } is
1 3 1 1 1 1 1 1 1 1 1 1 1 1 1
0, 1, , , + , + + , + + , · · · . Its sub sequence is , + , + +
2 4 2 8 2 8 16 2 8 32 2 2 8 2 8
∞
1 X 1
, · · · which are partial sums of the geometric series which converges
32 22n+1
n=0
2
and sum is given by . Thus in this case xn convereges to 32 .
3
xn − x1
Now in general, for any x1 < x2 , we define zn−1 = . Note that z0 = 0 and
x2 − x1
zn−1 + zn
z1 = 1 and the sequence zn satisfies the relation zn+1 = . Thus by the
2
special case above, zn converges to 23 . Now observe that the limit of xn is the real
number x which divides the interval [x1 , x2 ] in the ratio 2 : 1. For x = 0, we need
1
to take x1 6= 0 and x2 = − x1 . 6 marks
2
3. (a) Show that there does not exist a 3-digit number A such that 103 A + A is a perfect
square.
(b) Show that there exists an n-digit (n > 3) number A such that 10n A+A is a perfect
square. [12]
Solution:

(a) If 103 A + A = A(1001) is a perfect square, then 7|A, 11|A and 13|A. Therefore
A ≥ 1001. That is A has bigger than or equal to 4 digits. 3 marks
(b) For some n if 10n A + A = A(10n + 1) is a perfect square, then 10n + 1 must be
divisible by a square bigger than 1. Because if no perfect square divides 10n + 1,
then all prime divisors of 10n + 1 must appear in factorization of A. This makes
A ≥ 10n + 1, but A < 10n + 1. Therefore 10n + 1 has a square factor. 4 marks

4
 Now observe that 1011 +1 = (11)(1010 −109 +108 −· · ·−10+1) = (11)(9090909091)
and thus 121 divides 1011 + 1. 2 marks
1011 + 1 10 11 n
Thus we can choose A as × 9 so that 10 ≤ A < 10 and (10 + 1)A is
121
a perfect square. 3 marks

4. For n × n matrices A, B, let C = AB − BA. If C commutes with both A and B, then

(a) Show that AB k − B k A = kB k−1 C for every positive integer k.

(b) Show that there exists a positive integer m such that C m = 0. [13]

Solution:

(a) The proof is by induction. The result is true for k = 1, 2. AB − BA = C and

AB 2 − B 2 A = (AB − BA)B + B(AB − BA) = CB + BC = 2BC. 1 marks
Assume that the result is true upto k − 1. AB k − B k A = (AB − BA)B k−1 +
B(AB k−1 − B k−1 A) = CB k−1 + B(k − 1)B k−2 C = kB k−1 C. 3 marks
(b) Hence for any polynomial q(x), Aq(B) − q(B)A = q 0 (B)C, where q 0 is a derivative
of q. In particular, let p(x) be the characteristic polynomial of B. Now by Caley-
Hamilton Theorem,
0 = Ap(B) − p(B)A = p0 (B)C. This proves p0 (B)C = 0. 4 marks
0 = Ap0 (B)C−p0 (B)AC = (Ap0 (B)−p0 (B)A)C = p00 (B)C 2 . This proves p00 (B)C 2 =
0. Inductively, we have p(k) (B)C k = 0. Therefore for k = n, we have n!C n = 0.
Hence C n = 0. 5 marks