AMT P UBLISHING

101 PROBLEMS lN ALGEBRA

FROM THE TRAlNlNG OF THE USA lMO TEAM

T ANDREESClJ H Z FENG
 Published by

AMT PUBLISHING

Australian Mathematics Trust

University of Canberra ACT 2601
AUSTRAUA

Copyright ©2001 AMT Publishing

Telephone: +61 2 6201 5137

AMTT Limited ACN 083 950 341
National Library of Australia Card Number and ISSN
Australian Mathematics Trust Enrichment Series ISSN 1326-0170
101 Problems in Algebra ISBN 1 876420 12 X
THE AUSTRALIAN MATHEMATICS TRUST

ENRlCHMENT SERlES

ED1roR1'\1 Cof\1f\11nEE

• Chairman GRAHAM H POUARD, Canberra AUSTRALIA

• Editor PETER J TAYLOR, Canberra AUSTRALIA

WARREN J ATKINS, Canberra AUSTRALIA

ED J BARBEAU, Toronto CANADA
GEORGE BERZSENYl, Terra Haute USA
RON DUNKLEY, Waterloo CANADA
WALTER E MIENTKA, Lincoln USA
N1KOLAY KoNSTANTlNOV, Moscow Russ1A
ANDY Liu, Edmonton CANADA
JORDAN B TABOV, Sofia BULGARIA
JOHN WEBB, Cape Town SOUTH AFRICA

The books in this series are selected for their motivating, interesting
and stimulating sets of quality problems, with a lucid expository style
in their solutions. Typically, the problems have occurred in either
national or international contests at the secondary school level.
They are intended to be sufficiently detailed at an elementary level
for the mathematically inclined or interested to understand but, at
the same time, be interesting and sometimes challenging to the
undergraduate and the more advanced mathematician. 1t is believed
that these mathematics competition problems are a positive
influence on the learning and enrichment of mathematics.
 THE AUSTRALIAN MATHEMATICS TRUST

ENRlCHMENT SERlES

BOOl\'i IN I Ill Sl 1\11"

1 1 AUSTRALIAN MATHEMATICS COMPETITION BOOK 1 1978-1984

JD Edwards, DJ King Et PJ O'Halloran

1 2 MATHEMATICAL TOOLCHEST
AW Plank Et NH Williams
3 TOURNAMENT OF TOWNS QUESTIONS AND SOLUTIONS 1984-1989
PJ Taylor
4 AUSTRALIAN MATHEMATICS COMPE111lON BOOK 2 1985-1991
PJ O'Halloran, G Pollard Et PJ Taylor
5 PROBLEM SOLVING VIA THE AMC
W Atkins
6 TOURNAMENT OF TOWNS QUESTIONS AND SOLUTIONS 1980-1984
PJ Taylor
7 TOURNAMENT OF TOWNS QUESTIONS AND SOLUTIONS 1989-1993
PJ Taylor
8 ASIAN PACIFIC MATHEMATICS OLYMPIADS 1989-2000
H Lausch Et C Bosch Giral
I 9 METHODS OF PROBLEM SOLVING BOOK 1
JB Tabov Et PJ Taylor
I 10 CHALLENGE! 1991-1995
JB Henry, J Dawsey, AR Edwards, LI Mottershead,
A Nakos Et G Vardaro
I 11 USSR MATHEMATICAL OLYMPIADS 1989-1992
AM Slinko

1 12 AUSTRALIAN MATHEMATICAL OLYMPIADS 1979-1995

H Lausch Et PJ Taylor
1 13 CHINESE MATHEMATICS COMPETITIONS AND OLYMPIADS 1981-1993
A Liu
14 POLISH
1 ff AUSTRIAN MATHEMATICAL OLYMPIADS 1981- 1995
ME Kuczma Et E Windischbacher
15 TOURNAMENT OF TOWNS QUESTIONS AND SOLUTIONS 1993-1997
PJ Taylor Et AM Storozhev
,16 AUSTRALIAN MATHEMA11CS COMPETITION BOOK 3 1992-1998
WJ Atkins, JE Munro Et PJ Taylor
17
1 SEEKING SOLUTIONS
JC Burns
18 101 PROBLEMS IN ALGEBRA
T Andreescu Et Z Feng
 PREFACE

This book contains one hundred highly rated problems used in the train-
ing and testing of the USA International Mathematical Olympiad (IMO)
team. It is not a collection of one hundred very difficult, impenetrable
questions. Instead, the book gradually builds students' algebraic skills
and techniques. This work aims to broaden students' view of mathemat-
ics and better prepare them for possible participation in various mathe-
matical competitions. It provides in-depth enrichment in important areas
of algebra by reorganizing and enhancing students' problem-solving tac-
tics and strategies. The book further stimulates students' interest for
future study of mathematics.
 INTRODUCTION

In the United States of America, the selection process leading to par-

ticipation in the International Mathematical Olympiad (IMO) consists
of a series of national contests called the American Mathematics Con-
test 10 (AMC 10), the American Mathematics Contest 12 (AMC 12),
the American Invitational Mathematics Examination(AIME), and the
United States of America Mathematical Olympiad (USAMO). Partici-
pation in the AIME and the USAMO is by invitation only, based on
performance in the preceding exams of the sequence. The Mathemati-
cal Olympiad Summer Program (MOSP) is a four-week, intense train-
ing of 24-30 very promising students who have risen to the top of the
American Mathematics Competitions. The six students representing the
United States of America in the IMO are selected on the basis of their
USAMO scores and further IMO-type testing that takes place during
MOSP. Throughout MOSP, full days of classes and extensive problem
sets give students thorough preparation in several important areas of
mathematics. These topics include combinatorial arguments and identi-
ties, generating functions, graph theory, recursive relations, telescoping
sums and products, probability, number theory, polynomials, theory of
equations, complex numbers in geometry, algorithmic proofs, combinato-
rial and advanced geometry, functional equations and classical inequali-
ties.
Olympiad-style exams consist of several challenging essay problems. Cor-
rect solutions often require deep analysis and careful argument. Olym-
piad questions can seem impenetrable to the novice, yet most can be
solved with elementary high school mathematics techniques, cleverly ap-
plied.
Here is some advice for students who attempt the problems that follow.

• Take your time! Very few contestants can solve all the given prob-
lems.
• Try to make connections between problems. A very important
theme of this work is: all important techniques and ideas featured
in the book appear more than once!
• Olympiad problems don't "crack" immediately. Be patient. Try
different approaches. Experiment with simple cases. In some cases,
working backward from the desired result is helpful.
• Even if you can solve a problem, do read the solutions. They may
contain some ideas that did not occur in your solutions, and they
viii Introduction

may discuss strategic and tactical approaches that can be used else-
where. The formal solutions are also models of elegant presenta-
tion that you should emulate, but they often obscure the torturous
process of investigation, false starts, inspiration and attention to
detail that led to them. When you read the solutions, try to re-
construct the thinking that went into them. Ask yourself, "What
were the key ideas?" "How can I apply these ideas further?"
• Go back to the original problem later, and see if you can solve it
in a different way. Many of the problems have multiple solutions,
but not all are outlined here.
• All terms in boldface are defined in the Glossary. Use the glossary
and the reading list to further your mathematical education.
• Meaningful problem solving takes practice. Don't get discouraged
if you have trouble at first. For additional practice, use the books
on the reading list.
 ACKNOWLEDGEMENTS

Thanks to Tiankai Liu who helped in proof reading and preparing solu-
tions.
Many problems are either inspired by or fixed from mathematical contests
in different countries and from the following journals:
High-School Mathematics, China
Revista Matematicii. Timi§oara, Romania
Kvant, Russia

We did our best to cite all the original sources of the problems in the solu-
tion part. We express our deepest appreciation to the original proposers
of the problems.
 ABBREVIATIONS AND NOTATIONS

Abbreviations
AHS ME American High School Mathematics
Examination
AIME American Invitational Mathematics
Examination
AMClO American Mathematics Contest 10
AMC12 American Mathematics Contest 12,
which replaces AHSME
ARML American Regional Mathematics League
IMO International Mathematical Olympiad
USAMO United States of America Mathematical Olympiad
MOSP Mathematical Olympiad Summer Program
Putnam The William Lowell Putnam Mathematical
Competition
St. Petersburg St. Petersburg (Leningrad) Mathematical
Olympiad

Notations for Numerical Sets and Fields

z the set of integers
Zn the set of integers modulo n
N the set of positive integers
No the set of nonnegative integers
Q the set of rational numbers
Q+ the set of positive rational numbers
Qo the set of nonnegative rational numbers
Qn the set of n-tuples of rational numbers
~ JR the set of real numbers
JR+ the set of positive real numbers
JRO the set of nonnegative real numbers
Rn the set of n-tuples of real numbers
c the set of complex numbers
CONTENTS
• PREFACE vii

• INTRODUCTION ix

• ACKNOWLEDGEMENTS xi

• ABBREVIATIONS AND NOTATIONS xiii

• 1. INTRODUCTORY PROBLEMS

• 2. ADVANCED PROBLEMS 13

• 3. SOLUTIONS TO INTRODUCTORY PROBLEMS 27

• 4. SOLUTIONS TO ADVANCED PROBLEMS 65

• GLOSSARY 1 31

• FURTHER READING 137

lNTRODUCTORY PROBLEMS
 1. INTRODUCTORY PROBLEMS

Problem 1
Let a, b, and c be real and positive parameters. Solve the equation

Problem 2
Find the general term of the sequence defined by x 0 = 3, x 1 = 4 and

for all" n E N.

Problem 3
Let x 1 , x 2 , ... , Xn be a sequence of integers such that

(i) -1 :':: X 2 :':: 2, fori = 1, 2, ... , n;

(ii) X1 + X2 + · + Xn = 19;

(iii) x? + x§ + · · · + x; = 99.
Determine the minimum and maximum possible values of

Problem 4
The function f, defined by

f(x)=ax+b 1
ex+ d
where a, b, c, and d are nonzero real numbers, has the properties

!(19) = 19, f(97) = 97, and f(f(x)) = x,

d
for all values of x, except --.
c
Find the range of f.
2 1. Introductory Problems

Problem 5
Prove that
(a - b) 2 a+ b "b (a - b) 2
- - - < - - -vao< - - -
8a - 2 - 8b
for all a ;::: b > 0.

Problem 6
Several (at least two) nonzero numbers are written on a board. One may
erase any two numbers, say a and b, and then write the numbers a+~
and b - ~ instead.
Prove that the set of numbers on the board, after any number of the
preceding operations, cannot coincide with the initial set.

Problem 7
The polynomial

1- x + x2 - x3 + ... + xl6 - xl 7
may be written in the form

where y = x + 1 and ais are constants.

Find a2·

Problem 8
Let a, b, and c be distinct nonzero real numbers such that
1 1 1
a+-=b+-=c+-.
b c a
Prove that label = 1.

Problem 9
Find polynomials f(x), g(x), and h(x), if they exist, such that for all x,

-1 if x < -1
lf(x)l - lg(x)I + h(x) = { 3x + 2 if -1 ~ x ~ 0
-2x+2 if x > 0.
1. Introductory Problems 3

Problem 10
Find all real numbers x for which
gx + 27x 7
12x + l8X = 6'

Problem 11
Find the least positive integer m such that

(2:) ~ < m
for all positive integers n.

Problem 12
Let a, b, c, d, and e be positive integers such that

abcde = a + b + c + d + e.
Find the maximum possible value of max{ a, b, c, d, e}.

Problem 13
Evaluate
3 4 2001
1! + 2! + 3! + 2! + 3! + 4! + ... + 1999! + 2000! + 2001!.

Problem 14
Let x = .../a 2 +a+1 - .../a2 - a+ 1, a E JR.
Find all possible values of x.

Problem 15
Find all real numbers x for which
4 1. Introductory Problems

Problem 16
Let f: N x N ____, N be a function such that f(l, 1) = 2,

f(m+ 1,n) = f(m,n) +m and f(m,n+ 1) = f(m,n)-n

for all m, n E N.
Find all pairs (p, q) such that f(p, q) = 2001.

Problem 17
Let f be a function defined on [O, 1] such that

f(O) = /(1) = 1 and IJ(a) - f(b)I < la - bl,

for all a i= b in the interval [O, l].

Prove that
1
lf(a) - f(b)I < 2.
Problem 18
Find all pairs of integers ( x, y) such that

x3 + Y3 = (x + Y )2.

Problem 19
2
Let f(x) =- - for real numbers x.
4x + 2
Evaluate
1 ( 20101 ) 2
+ / ( 20 01 ) + · · · + / G~~~) ·
Problem 20
Prove that for n ~ 6 the equation
1 1 1
-+-+···+-=1
xi x~ x;

has integer solutions.

Problem 21
Find all pairs of integers (a, b) such that the polynomial ax 17 + bx 16 + 1
is divisible by x 2 - x - l.
1. Introductory Problems 5

Problem 22
Given a positive integer n, let p(n) be the product of the non-zero digits
of n. (If n has only one digit, then p(n) is equal to that digit.) Let

s = p(l) + p(2) + ... + p(999).

What is the largest prime factor of S?

Problem 23
Let Xn be a sequence of nonzero real numbers such that

for n = 3,4, ....

Establish necessary and sufficient conditions on x1 and X2 for Xv- to be
an integer for infinitely many values of n.

Problem 24
Solve the equation
x3 - 3x = ..Jx + 2.

Problem 25
For any sequence of real numbers A = { ai, a2, aa, · · ·}, define .6.A to be
the sequence {a2 - ai, aa - a2, a4 - aa, ... }. Suppose that all of the terms
of the sequence .6.(.6.A) are 1, and that a19 = a92 = 0.
Find ai.

Problem 26
Find all real numbers x satisfying the equation
2x + 3x _ 4x + 6x _ gx = 1.

Problem 27
Prove that
80 1
16 < I: vk < 11.
k=l
/1"

Problem 28
Determine the number of ordered pairs of integers (m, n) for which mn ~
0 and
m 3 + n 3 + 99mn = 333 .
6 1. Introductory Problems

Problem 29
Let a, b, and c be positive real numbers such that a + b+ c ~ 4 and
ab + be+ ca ~ 4.
Prove that at least two of the inequalities

la - bl ~ 2, lb - cl ~ 2, le - al ~ 2

are true.

Problem 30
Evaluate
n 1
~ (n - k)!(n + k)! ·

Problem 31
Let 0 < a < 1. Solve

for positive numbers x.

Problem 32
What is the coefficient of x 2 when

(1 + x)(l + 2x)(l + 4x) · · · (1 + 2nx)

is expanded?

Problem 33
Let m and n be distinct positive integers.
Find the maximum value of lxm - xnl, where x is a real number in the
interval (0, 1).

Problem 34
Prove that the polynomial

where a 1 , a 2 , · · · , an are distinct integers, cannot be written as the prod-

uct of two non-constant polynomials with integer coefficients, i.e., it is
irreducible.
1. Introductory Problems 7

Problem 35
Find all ordered pairs of real numbers (x, y) for which:

(1 + x)(l + x 2 )(1 + x4 )
and (1 + y)(l + y 2 )(1 + y4 )

Problem 36
Solve the equation

2(2x - l)x 2 + (2x 2 - 2)x = 2x+l - 2

for real numbers x.

Problem 37
Let a be an irrational number and let n be an integer greater than 1.
Prove that

is an irrational number.

Problem 38
Solve the system of equations

(x1 - X2 + X3) 2 X2(X4 + X5 - X2)

(x2 - X3 + X4) 2 X3(X5 + X1 - X3)

(x3 - X4 + xs) 2 X4(X1 + X2 - X4)

(x4 - X5 + X1) 2 X5(X2 + X3 - X5)

(xs - x1 + x2) 2 x1(x3 + X4 - x1)

Problem 39
Let x, y, and z be complex numbers such that

x+y+z = 2,
x2 + y2 + z2 = 3
and
xyz = 4.
Evaluate
1 1 1
----+ +----
xy + z - 1 yz + x - 1 zx + y - 1
8 1. Introductory Problems

Problem 40
Mr. Fat is going to pick three non-zero real numbers and Mr. Taf is going
to arrange the three numbers as the coefficients of a quadratic equation

_x 2 + _x + _ = 0.
Mr. Fat wins the game if and only if the resulting equation has two
distinct rational solutions.
Who has a winning strategy?

Problem 41
Given that the real numbers a, b, c, d, and e satisfy simultaneously the
relations

a+ b + c + d + e = 8 and a 2 + b2 + c2 + d 2 + e2 = 16,
determine the maximum and the minimum value of a.

Problem 42
Find the real zeros of the polynomial

Pa(x) = (x 2 + l)(x - 1) 2 - ax 2 ,

where a is a given real number.

Problem 43
Prove that
1 3 2n - 1 1
-·-···--<--
2 4 2n ffn
for all positive integers n.

Problem 44
Let
P(x) = aoxn + a1xn-l +···+an
be a nonzero polynomial with integer coefficients such that P(r)
P(s) = 0 for some integers rand s, with 0 < r < s.
Prove that ak ::; -s for some k.

Problem 45
Let m be a given real number.
Find all complex numbers x such that
1. Introductory Problems 9

Problem 46
The sequence given by Xo = a, X1 = b, and

is periodic.
Prove that ab= l.

Problem 47
Let a, b, c, and d be real numbers such that

(a 2 + b2 - l)(c2 + d2 - 1) > (ac + bd - 1)2.

Prove that
a 2 + b2 > 1 and c2 + d2 > l.

Problem 48
Find all complex numbers z such that

(3z + 1)(4z + 1)(6z + 1)(12z + 1) = 2.

Problem 49
Let xi, x2, .. ·, Xn- 1, be the zeros different from 1 of the polynomial
P(x) = xn - 1, n ~ 2.
Prove that
1 1 1 n-1
- -+-
l-x1
- + .. ·+ 1-Xn-1
l-x2
=-2-·

Problem 50
Let a and b be given real numbers. Solve the system of equations

x -yJx2 -y2
= a,
Jl - x2 +y2
y-xJx2 -y2
Jl -x2 +y2 = b

for real numbers x and y.

ADVANCED PROBLEMS
 2. ADVANCED PROBLEMS

Problem 51
Evaluate

(2000)
2 + (2000)
5 + (2000)
8 + ...+ (2000)
2000 .
Problem 52
Let x, y, z be positive real numbers such that x 4 + y4 + z 4 = 1.
Determine with proof the minimum value of
x3 y3 z3
- - +1-- -y 8+1-
1 - x8
-
- z8 ·

Problem 53
Find all real solutions to the equation

Problem 54
Let {an}n;::::1 be a sequence such that ai = 2 and

for all n EN.

Find an explicit formula for an.

Problem 55
Let x, y, and z be positive real numbers. Prove that

x + ---;===y==;:==:::;::
x+ J(x+y)(x+z) y+ J(y+z)(y+x)

+ z < 1.
z + y'(z + x)(z + y)
14 2. Advanced Problems

Problem 56
Find, with proof, all nonzero polynomials f (z) such that
J(z 2 ) + J(z)f(z + 1) = 0.

Problem 57
Let f: N---> N be a function such that J(n + 1) > J(n) and
J(J(n)) = 3n

for all n.
Evaluate /(2001).

Problem 58
Let F be the set of all polynomials f(x) with integers coefficients such
that J(x) = 1 has at least one integer root.
For each integer k > 1, find mk, the least integer greater than 1 for
which there exists f E F such that the equation f (x) = mk has exactly
k distinct integer roots.

Problem 59
Let x 1 = 2 and
Xn+I = x; - Xn + 1,
for n :'.'.'. l.
Prove that
1 1 1 1 1
1- - - < -
22 n - l XI
+-
X2
+ ... + -Xn < 1- -
22
.
n

Problem 60
Suppose that J : JR+ ---> JR+ is a decreasing function such that for all
x, y E JR+,

J(x + y) + J(J(x) + J(y)) = J(J(x + J(y)) + J(y + J(x))).

Prove that J(J(x)) = x.
2. Advanced Problems 15

Problem 61
Find all functions f : Ql --t Ql such that
f(x + y) + f(x - y) = 2/(x) + 2/(y)
for all x, y E Ql.

Problem 62
Let~< a< 1.
Prove that the equation

x 3 (x + 1) = (x + a)(2x +a)

has four distinct real solutions and find these solutions in explicit form.

Problem 63
Let a, b, and c be positive real numbers such that abc = 1.
Prove that
1 1 1
---+
a+b+l
+
b+c+l
<1.
c+a+l-

Problem 64
Find all functions f, defined on the set of ordered pairs of positive inte-
gers, satisfying the following properties:

f(x, x) = x, f(x, y) = f(y, x), (x + y)f(x, y) = yf(x, x + y).

Problem 65
Consider n complex numbers zk, such that lzkl ::; 1, k = 1, 2, ... , n.
Prove that there exist ei, e2, ... , en E {-1, 1} such that, for any m ::; n,

Problem 66
Find a triple of rational numbers (a, b, c) such that

«-~ 1= ?la+ w+ ~.
16 2. Advanced Problems

Problem 67
Find the minimum of

where x1, x2, ... , Xn are real numbers in the interval (i, 1).
Problem 68
Determine x 2 + y 2 + z 2 + w 2 if
x2 y2 z2 w2
22 - 12 + 22 - 32 + 22 - 52 + 22 - 72 = l,
x2 y2 z2 w2
42 - 12 + 42 - 32 + 42 - 52 + 42 - 72 = l,

x2 y2 z2 w2
62 - 12 + 62 - 32 + 62 - 52 + 62 - 72 = l,

x2 y2 z2 w2
82 - 12 + 82 - 32 + 82 - 52 + 82 - 72 = 1.

Problem 69
Find all functions f : JR --+ JR such that

f(xf(x) + f(y)) = (f(x)) 2 +y

for all x, y E JR.

Problem 70
The numbers 1000, 1001, · · ·, 2999 have been written on a board.
Each time, one is allowed to erase two numbers, say, a and b, and replace
them by the number ~min( a, b).
After 1999 such operations, one obtains exactly one number c on the
board. Prove that c < 1.

Problem 71
Let a 1 , a 2 , ... , an be real numbers, not ail zero.
Prove that the equation

v'l + aix + v'l + a2x + · · · + v'l + anx = n

has at most one nonzero real root.
2. Advanced Problems 17

Problem 72
Let {an} be the sequence of real numbers defined by a 1 = t and

an+l = 4an(l - an)

for n ~ 1.
For how many distinct values of t do we have a 1998 = O?

Problem 73

(a) Do there exist functions f: JR--+ JR and g : JR--+ JR such that

f(g(x)) = x 2 and g(f(x)) = x 3
for all x E JR?
(b) Do there exist functions f : JR --+ JR and g : JR --+ JR such that
f(g(x)) = x 2 and g(f(x)) = x4
for all x E JR?

Problem 74
Let 0 < a1 :5 a2 · · · :5 an, 0 < b1 :5 b2 · · · :5 bn be real numbers such that
n n
I.:ai ~ 2:.:bi.
i=l i=l

Suppose that there exists 1 :5 k :5 n such that bi :5 ai for 1 :5 i :5 k and

bi ~ ai for i > k.
Prove that

Problem 75
Given eight non-zero real numbers a 1 , a 2 , · · ·, a8 , prove that at least one
of the following six numbers: a1a3 + a2a4, a1a5 + a2a6, a1a1 + a2aa,
a3a5 + a4a6, a3a1 + a4aa, a5a1 + a5aa is non-negative.

Problem 76
Let a, b and c be positive real numbers such that abc = 1.
Prove that
~ ~
---5--+ + 00
< 1.
~+b+~ ~+~+~ ~+~+ro-
18 2. Advanced Problems

Problem 77
Find all functions f : JR. --+ JR. such that the equality

f(f(x) + y) = f(x 2 - y) + 4/(x)y

holds for all pairs of real numbers (x, y).

Problem 78
Solve the system of equations:
3x -y
x+ 2
x +y
2 =3

- x +3y - 0
Y x2 + y2 - ·

Problem 79
Mr. Fat and Mr. Taf play a game with a polynomial of degree at least 4:

x2n + _x2n-l + _x2n-2 + ... + _x + 1.

They fill in real numbers to empty spaces in turn. If the resulting poly-
nomial has no real root, Mr. Fat wins; otherwise, Mr. Taf wins.
If Mr. Fat goes first, who has a winning strategy?

Problem 80
Find all positive integers k for which the following statement is true: if
F(x) is a polynomial with integer coefficients satisfying the condition

0 ~ F(c) ~ k for c = 0, 1, .. ., k + 1,
then F(O) = F(l) = · · · = F(k + 1).
Problem 81
The Fibonacci sequence Fn is given by

Prove that

for all n ~ 2.
2. Advanced Problems 19

Problem 82
Find all functions u : JR --+ JR for which there exists a strictly monotonic
function f : JR --+ JR such tha~

f(x + y) = f(x)u(y) + f(y)

for all x, y E JR.

Problem 83
Let zi, z2, . .. , Zn be complex numbers such that

Prove that there exists a subset S of {zi, z 2, ... , Zn} such that

Problem 84
A polynomial P(x) of degree n ;::: 5 with integer coefficients and n distinct
integer roots is given.
Find all integer roots of P(P(x)) given that 0 is a root of P(x).

Problem 85
Two real sequences x 1 , x2, ... , and y 1 , Y2, ... , are defined in the following
way:
X1 =YI= J3, Xn+I = Xn + Jl + x;,
and
Yn
Yn+ i = l
+ Vlfl-:::2
+Yr.2
.i

for all n ;::: 1. Prove that 2 < Xn]Jn < 3 for all n > 1.

Problem 86
For a polynomial P(x), define the difference of P(x) on the interval [a, b]
([a, b), (a, b), (a, b]) as P(b) - P(a).
Prove that it is possible to dissect the interval [O, 1] into a finite number
of intervals and color them red and blue alternately such that, for every
quadratic polynomial P(x), the total difference of P(x) on red intervals
is equal to that of P(x) on blue intervals.
What about cubic polynomials?
20 2. Advanced Problems

Problem 87
Given a cubic equation

x3 + _x 2 + _x + _ = 0,
Mr. Fat and Mr. Taf are playing the following game. In one move, Mr.
Fat chooses a real number and Mr. Taf puts it in one of the empty spaces.
After three moves the game is over. Mr. Fat wins the game if the final
equation has three distinct integer roots.
Who has a winning strategy?

Problem 88
Let n > 2 be an integer and let f : JR. 2 -> JR be a function such that for
any regular n-gon A1 A2 ... An,

/(Ai)+ J(A2) +···+/(An) = 0.

Prove that f is the zero function.

Problem 89
Let p be a prime number and let f(x) be a polynomial of degreed with
integer coefficients such that:

(i) /(0) = 0, /(1) = 1;

(ii) for every positive integer n, the remainder upon division of f(n)
by p is either 0 or l.
Prove that d 2'. p - l.

Problem 90
Let n be a given positive integer.
Consider the sequence a0 , a 1 , ···,an with a0 = ~ and

for k = 1, 2, · · · , n.
Prove that
1
l--<an<l.
n
2. Advanced Problems 21

Problem 91
Let a 1 , a 2 , ..• , an be nonnegative real numbers, not all zero.
(a) Prove that xn - a1xn-I - · · · - an-IX - an = 0 has precisely one
positive real root R.

(b) Let A= L:;=l ai and B = L:;=l 1a1 .

Prove that AA :::; R 8 .

Problem 92
Prove that there exists a polynomial P(x, y) with real coefficients such
that P(x, y) ;::: 0 for all real numbers x and y, which cannot be written
as the sum of squares of polynomials with real coefficients.

Problem 93
For each positive integer n, show that there exists a positive integer k
such that
k = f(x)(x + 1) 2 n + g(x)(x 2n + 1)
for some polynomials f, g with integer coefficients, and find the smallest
such k as a function of n.

Problem 94
Let x be a positive real number.
(a) Prove that
00
(n-1)! 1
~ (x + 1) .. · (x + n) = -;·

(b) Prove that

00
(n - 1)! 00
1
L
n=l n(x + 1) ... (x + n) =
• L (x + k)2 ·
k=l
22 2. Advanced Problems

Problem 95
Let n 2: 3 be an integer, and let

X ~ S = {1, 2, ... , n 3 }
be a set of 3n 2 elements.
Prove that one can find nine distinct numbers a., b., c, (i = 1, 2, 3) in X
such that the system

a1x + b1y + c1z 0

a2x + b2y + c2z 0
aJX + b3y + C3Z 0

has a solution (xo, yo, zo) in nonzero integers.

Problem 96
Let n 2: 3 be an integer and let x 1 , x 2 , · · ·, Xn be positive real numbers.
n 1
Suppose that I:--= l.
1 +x
J=l 3

Prove that

1 1 1 )
vlxl+JX2+···+y'x;;°2:(n-1) ( -+-+···+-
Fi VX2 ..;x;; .
Problem 97
Let x 1 , x 2 , ... , Xn be distinct real numbers. Define the polynomials

P(x) = (x - x1)(x - x2) · · · (x - Xn)

and
1 1 1 )
Q(x)=P(x) ( --+--+···+-- .
X - X1 X - X2 X - Xn

Let Y1, Y2, ... , Yn-1 be the roots of Q. Show that

2. Advanced Problems 23

Problem 98
Show that for any positive integer n, the polynomial

f(x) = (x 2 + x) 2 n + 1
cannot be written as the product of two non-constant polynomials with
integer coefficients.

Problem 99
Let Ji, h, h : JR ___. JR be functions such that

is monotonic for all a1, a2, a3 E JR.

Prove that there exist c1 , c2 , c3 E JR, not all zero, such that

for all x ER

Problem 100
Let X1, X2, ... , Xn be variables, and let Y1, Y2, ... , Y2n -1 be the sums of
nonempty subsets of Xi·
Let Pk(x 1 , ... , xn) be the kth elementary symmetric polynomial in
the Yi (the sum of every product of k distinct Yis).
For which k and n is every coefficient of Pk (as a polynomial in x1, ... , Xn)
even?
For example, if n = 2, then Y1, Y2, y3 are x1, x2, x1 + x2 and

P1 = Y1 + Y2 + Y3 = 2x1 + 2x2,
P2 = Y1Y2 + Y2Y3 + Y3Y1 = xf + x~ + 3x1x2,
2
p3 = Y1Y2Y3 = X1X2 + X1X2·2

Problem 101
Prove that there exist 10 distinct real numbers a1, a2, ... , a10 such that
the equation

(x - a1)(x - a2) · · · (x - a10) = (x + a1)(x + a2) · · · (x + a10)

has exactly 5 different real roots.
SOLUTlONS TO
lNTRODUCTORY PROBLEMS
 3. SOLUTIONS TO
INTRODUCTORY PROBLEMS

Problem 1 [Romania 1974]

Let a, b, and c be real and positive parameters.
Solve the equation

.,/a+ bx+ ../b +ex+ .,/c +ax= vb - ax+ .,/c - bx+ .,/a - ex.

Solution 1
It is easy to see that x = 0 is a solution. Since the right hand side is a
decreasing function of x and the left hand side is an increasing function
of x, there is at most one solution.
Thus x = 0 is the only solution to the equation.

Problem 2
Find the general term of the sequence defined by Xo = 3, X1 = 4 and

for all n EN.

Solution 2
We shall prove by induction that Xn = n + 3. The claim is evident for
n = 0, 1.
For k ~ 1, if Xk-1 = k + 2 and Xk = k + 3, then
Xk+1 = xL 1 - kxk = (k + 2) 2 - k(k + 3) = k + 4,
as desired.
This completes the induction.
28 3. Solutions to Introductory Problems

Problem 3 [AHSME 1999]

Let x 1 , x 2 , ... , Xn be a sequence of integers such that
(i) -1 ~ Xi ~ 2, for i = 1, 2, ... , n;

(ii) X1+ X2 + · · · + Xn = 19;

(iii) x~ + x~ + · · · + x; = 99.
Determine the minimum and maximum possible values of

Solution 3
Let a, b, and c denote the number of - ls, ls, and 2s in the sequence,
respectively. We need not consider the zeros. Then a, b, care nonnegative
integers satisfying

-a + b + 2c = 19 and a + b + 4c = 99.
It follows that a= 40-c and b = 59-3c, where 0 ~ c ~ 19 (since b;::: 0),
so
x~ + x~ + · · · + x! = -a + b + Be = 19 + 6c.
When c = 0 (a= 40, b = 59), the lower bound (19) is achieved.
When c = 19 (a= 21, b = 2), the upper bound (133) is achieved.

Problem 4 [AIME 1997]

The function/, defined by

f(x) = ax+b,
cx+d
where a, b, c, and d are nonzero real numbers, has the properties

/(19) = 19, /(97) = 97, and f(f(x)) = x,

d
for all values of x, except--.
c
Find the range of f.
Solution 4, Alternative 1
For all x, f(f(x)) = x, i.e.,

a (ax+b) +b
cx+d
c (ax+ b) +d = x,
cx+d
3. Solutions to Introductory Problems 29

i.e.
(a 2 + bc)x + b(a + d)
---------x
c(a+d)x+bc+d2 - '
i.e.
c(a + d)x 2 + (d 2 - a2)x - b(a + d) = 0,
which implies that c( a + d) = 0. Since c =/; 0, we must have a = -d.
The conditions /(19) = 19 and /(97) = 97 lead to the equations

192 c = 2 · l9a + b and 972 c = 2 · 97a + b.

Hence
(972 - 192 )c = 2(97 - 19)a.
It follows that a = 58c, which in turn leads to b = -1843c. Therefore

f(x) = 58x - 1843 = 58 + 1521

x - 58 x - 58'
which never has the value 58.
Thus the range off is JR. - {58}.
Solution 4, Alternative 2
The statement implies that f is its own inverse. The inverse may be
found by solving the equation
ay+b
x=--
cy+d
for y. This yields
1-1(x)= dx-b.
-cx+a
The nonzero numbers a, b, c, and d must therefore be proportional to d,
-b, -c, and a, respectively; it follows that a = -d, and the rest is the
same as in the first solution.

Problem 5
Prove that
(a - b) 2 a+ b r;-b (a - b) 2
-'------'- < - - - v ao < -'-----'-
Ba -2 - 8b
for all a ~ b > 0.
Solution 5, Alternative 1
Note that

( ya+
2ya
y'b) 2 < 1 < (ya+
- - 2v'b
y'b) 2
,
30 3. Solutions to Introductory Problems

i.e.

i.e.
(a-b) 2 < a-2Jlib+b < (a-b) 2
8a - 2 8b '
from which the result follows.
Solution 5, Alternative 2
Note that
2
( a+b ) b
a+ b _ Jlib = -2- - a (a - b) 2
2 a+b
--+vaor-;-b 2(a + b) + 4y'lib'
2
Thus the desired inequality is equivalent to

4a ? a + b + 2Vab ? 4b,

which is evident as a? b > 0 (which implies a? Jlib? b).

Problem 6 [St. Petersburg 1989]

Several (at least two) nonzero numbers are written on a board. One may
erase any two numbers, say a and b, and then write the numbers a +~
and b - ~ instead.
Prove that the set of numbers on the board, after any number of the
preceding operations, cannot coincide with the initial set.
Solution 6
Let S be the sum of the squares of the numbers on the board. Note that
S increases in the first operation and does not decrease in any successive
operation, as

with equality only if a = b = 0.

This completes the proof.
3. Solutions to Introductory Problems 31

Problem 7 [AIME 1986]

The polynomial
1 - x + x2 - x3 + ... + xl6 - x11

may be written in the form

ao + a1y + a2y 2 + · · · + a 16 y 16 + a17y17 ,

where y = x + 1 and ais are constants. Find a 2 .
Solution 7, Alternative 1
Let f(x) denote the given expression. Then

x !( x ) =x-x 2 +x 3 -···-x 18
and
(1+x)f(x)=1- x 18 .
Hence
f(x) = f(y - 1) = 1 - (y - 1)1s 1 - (y - 1)1s
l+(y-1) y
Therefore a 2 is equal to the coefficient of y 3 in the expansion of

1 - (y - 1)18'
i.e.,

Solution 7, Alternative 2
Let f(x) denote the given expression. Then

f(x) = f(y - 1) = 1 - (y - 1) + (y - 1) 2 - ... - (y - 1) 17

= 1 + (1- y) + (1 - y) 2 + ... + (1 - y) 17 .
Thus

Here we used the formula

and the fact that

G) = G) = l.
32 3. Solutions to Introductory Problems

Problem 8
Let a, b, and c be distinct nonzero real numbers such that
1 1 1
a+ - = b + - = c + -.
b c a
Prove that label = 1.
Solution 8
From the given conditions it follows that
b-c c-a a-b
a - b= --, b- c = --, and c - a = --.
be ca ab
Multiplying the above equations gives (abc) 2 = 1, from which the desired
result follows.

Problem 9 [Putnam 1999]

Find polynomials f(x), g(x), and h(x), if they exist, such that for all x,
-1 if x < -1
l/(x)l - lg(x)I + h(x) = { 3x + 2 if -1 :=:; x :=:; 0
-2x + 2 if x > 0.

Solution 9, Alternative 1
Since x = -1 and x = 0 are the two critical values of the absolute
functions, one can suppose that
F(x) alx + ll + blxl +ex+ d
(c-a-b)x+d-a ifx<-1
= { (a+c-b)x+a+d if-l:=:;x:=:;O
(a+ b + c)x +a+ d if x > 0,
which implies that a= 3/2, b = -5/2, c = -1, and d = 1/2.
Hence f(x) = (3x + 3)/2, g(x) = 5x/2, and h(x) = -x + ~­
Solution 9, Alternative 2
Note that if r(x) and s(x) are any two functions, then

max (r,s ) =
r + s + Ir - sl
.
2
Therefore, if F(x) is the given function, we have
F(x) max{-3x - 3, O} - max{5x, O} + 3x + 2
(-3x - 3 + l3x + 31)/2 - (5x + l5xl)/2 + 3x + 2
1
l(3x + 3)/2l - l5x/21- x + 2·
3. Solutions to Introductory Problems 33

Problem 10
Find all real numbers x for which
Bx+ 27x 7
12x+ 1sx = 6'

Solution 10
By setting 2x = a and 3x = b, the equation becomes

a 3 + b3 7
a 2 b + b2 a 6'
i.e.
a2 - ab+ b2 7
ab - 6'
i.e.
6a 2 - 13ab + 6b 2 = 0,
i.e.
(2a - 3b)(3a - 2b) = 0.
Therefore 2x+l = 3x+I or 2x-l = 3x- 1 , which implies that x = -1 and
x = 1.
It is easy to check that both x = -1 and x = 1 satisfy the given equation.

Problem 11 [Romania 1990]

Find the least positive integer m such that

(2:) ~ < m

for all positive integers n.

Solution 11
Note that

(2:) < (2;) + (2;) + ... + G~) = (1+1)2n = 4n

and for n = 5,
(150) = 252 > 35.
Thus m = 4.
34 3. Solutions to Introductory Problems

Problem 12
Let a, b, c, d, and e be positive integers such that

abcde = a + b + c + d + e.

Find the maximum possible value of max{ a, b, c, d, e }.

Solution 12, Alternative 1
Suppose that a ~ b ~ c ~ d ~ e. We need to find the maximum value of
e. Since
e < a + b + c + d + e ~ 5e,
then e < abcde ~ 5e, i.e. 1 < abed ~ 5.
Hence (a, b, c, d) = (1, 1, 1, 2), (1, 1, 1, 3), (1, 1, 1, 4), (1, 1, 2, 2), or
(1, 1, 1,5), which leads to max{e} = 5.
Solution 12, Alternative 2
As before, suppose that a ~ b~ c ~ d ~ e. Note that
1 1 1 1 1
l=-+-+-+-+-
bcde cdea deab eabc abed

< _!._+_!._+_!._+!+! = 3+d+e

- de de de e d de
Therefore, de~ 3 + d + e or (d - l)(e - 1) ~ 4.
If d = 1, then a = b = c = 1 and 4 + e = e, which is impossible.
Thus d - 1 ~ 1 and e - 1 ~ 4 or e ~ 5.
It is easy to see that (1, 1, 1, 2, 5) is a solution.
Therefore max{ e} = 5.

Comment: The second solution can be used to determine the maxi-

mum value of {x1, x2, ... , Xn}, when x1, x2, ... , Xn are positive integers
such that

Problem 13
Evaluate
3 4 2001
1! + 2! + 3! + 2! + 3! + 4! + ... + 1999! + 2000! + 2001!.
3. Solutions to Introductory Problems 35

Solution 13
Note that
k+2 k+2
=
k! + (k + 1)! + (k + 2)! k![l + k + 1 + (k + l)(k + 2)]

1
=
k!(k + 2)
k+l
(k + 2)!

(k + 2) - 1
(k + 2)!

1 1
= (k + 1)! (k+2)!'

By telescoping sum, the desired value is equal to

1 1
2- 2001!'

Problem 14
Let x = ../a2 +a+ 1- ../a2 - a+ 1, a ER
Find all possible values of x.
Solution 14, Alternative 1
Since

and
2a
x= '
../a2 +a+ 1 + ../a2 - a+ 1
we have
lxl < l2a/al = 2.

Squaring both sides of

x + J a2 - a + 1 = J a 2 + a + 1

yields
2xJa2 - a+ 1 = 2a - x 2.
36 3. Solutions to Introductory Problems

Squaring both sides of the above equation gives

· x 2 (x 2 - 4)
4(x 2 - l)a 2 = x 2 (x 2 - 4) or a 2 = .
4(x 2 - 1)

Since a 2 ;::: 0, we must have

x 2 (x 2 - 4)(x 2 - 1) ;::: 0,

Since !xi < 2, x 2 -4 < 0 which forces x 2 -1 < 0. Therefore, -1 < x < 1.
Conversely, for every x E (-1, 1) there exists a real number a such that

x= J a 2 +a+ 1 - J a 2 - a+ 1.

Solution 14, Alternative 2

Let A = (-1/2, ./3/2), B = (1/2, ./3/2), and P = (a, 0). Then P
is a point on the x-axis and we are looking for all possible values of
d=PA-PB.
By the Triangle Inequality, IPA- PBI < IABI = 1. And it is clear
that all the values -1 < d < 1 are indeed obtainable. In fact, for such
a d, a half hyperbola of all points Q such that QA - QB = d is well
defined. (Points A and B are foci of the hyperbola.)
Since line AB is parallel to the x-axis, this half hyperbola intersects the
x- axis, i.e., P is well defined.

Problem 15
Find all real numbers x for which

Solution 15
It is easy to check that x = 2 is a solution. We claim that it is the only
one. In fact, dividing by 13x on both sides gives

( 13lO)x + (.!.!)x
13
(12)x = l
+ 13
(14)x
+ 13
The left hand side is a decreasing function of x and the right hand side
is an increasing function of x.
Therefore their graphs can have at most one point of intersection.
3. Solutions to Introductory Problems 37

Comment: More generally,

a 2 +(a+ 1) 2 +···+(a+ k) 2
= (a+ k + 1) 2 +(a+ k + 2) 2 + · · · +(a+ 2k) 2
for a= k(2k + 1), k EN.
Problem 16 [Korean Mathematics Competition 2001]
Let f: N x N ~ N be a function such that f(l, 1) = 2,

f(m + 1, n) = f(m, n) +m and f(m, n + 1) = f(m, n) - n

for all m, n E N.
Find all pairs (p, q) such that f(p, q) = 2001.
Solution 16
We have

f (p, q) = f (p - 1, q) +p - 1
f(p- 2,q) + (p- 2) + (p-1)
=
= f(l,q) + p(p; l)
p(p- 1)
= f(l, q - 1) - (q - 1) + 2

f (1, 1) - q(q; l) + p(p; l)

2001.

Therefore
p(p- l) - q(q- l) = 1999
2 2 '
i.e.
(p - q)(p + q - 1) = 2. 1999.

Note that 1999 is a prime number and that p-q < p+q-1 for p,q EN.
We have the following two cases:

1. p - q = 1 and p + q - 1 = 3998. Hence p = 2000 and q = 1999.

2. p - q = 2 and p + q - 1 = 1999. Hence p = 1001 and q = 999.

38 3. Solutions to Introductory Problems

Therefore (p, q) = (2000, 1999) or (1001, 999).

Problem 17 [China 1983]

Let f be a function defined on [O, 1] such that

/(0) = /(1) = 1 and l/(a) - /(b)I <la - bl,

for all a-=/:- bin the interval [O, l].
Prove that
1
lf(a) - /(b)I < 2.
Solution 17
We consider the following cases.

1. la - bl:::; 1/2. Then l/(a) - /(b)I <la - bl:::;~' as desired.

2. la - bl > 1/2. By symmetry, we may assume that a > b. Then
l/(a) - /(b)I l/(a) - /(1) + /(0) - /(b)I
< lf(a) - /(1)1 + l/(o) - /(b)I
< la-ll+IO-bl
1-a+b-O
1- (a - b)
1
< 2'
as desired.

Problem 18
Find all pairs of integers (x, y) such that

x3 + y3 = (x + y)2.

Solution 18
Since x 3 +y 3 = (x + y)(x 2 - xy + y 2 ), all pairs of integers (n, -n), n E Z,
are solutions.
Suppose that x + y -=/:- 0. Then the equation becomes

x2 - xy + y 2 = x + y,
i.e.
x2 - (y + l)x + y 2 - y = 0.
3. Solutions to Introductory Problems 39

Treated as a quadratic equation in x, we calculate the discriminant

b. = y2 + 2y + 1 - 4y2 + 4y = -3y 2 + 6y + 1.

Solving for b. ;::: 0 yields

3-2J3 3+2J3
3 5,.y5:_ 3 .

Thus the possible values for y are 0, 1, and 2, which lead to the solutions
(1, 0), (0, 1), (1, 2), (2, 1), and (2, 2).
Therefore, the integer solutions of the equation are (x, y) = (1, 0), (0, 1),
(1, 2), (2, 1), (2, 2), and (n, -n), for all n E Z.

Problem 19 [Korean Mathematics Competition 2001]

Let
2
f(x) = 4x +2
for real numbers x. Evaluate

I (20~1) +I (2~01) +···+I(~~~~)·

Solution 19
Note that f has a half-turn symmetry about point (1/2, 1/2). Indeed,
2
/(1 - x) = 4-1--x-+-2

from which it follows that f(x) + /(1 - x) = 1.

Thus the desired sum is equal to 1000.

Problem 20
Prove that for n ;::: 6 the equation
1 1 1
-+-+···+-=1
x~ x~ x;
has integer solutions.
Solution 20
Note that
1 1 1 1 1
a 2 = (2a)2 + (2a) 2 + (2a) 2 + (2a) 2 '
40 3. Solutions to Introductory Problems

from which it follows that if (x 1 , x 2 , · · ·, Xn) = (a 1 , a 2 , ···,an) is an inte-

ger solution to
1 1 1
-
x21+ - x22+ .. · +x2 - -- 1 '
n

then

(x1, X2, · · ·, Xn-1, Xn, Xn+l> Xn+2, Xn+3)

= (a1, a2, · · ·, an-1, 2an, 2an, 2an, 2an,)
is an integer solution to

Therefore we can construct the solutions inductively if there are solutions

for n = 6, 7, and 8.
Since x 1 = 1 is a solution for n = 1, (2, 2, 2, 2) is a solution for n = 4,
and (2, 2, 2, 4, 4, 4, 4) is a solution for n = 7.
It is easy to check that (2, 2, 2, 3, 3, 6) and (2, 2, 2, 3, 4, 4, 12, 12) are solu-
tions for n = 6 and n = 8, respectively. This completes the proof.

Problem 21 [AIME 1988]

Find all pairs of integers (a, b) such that the polynomial

ax 17 + bx 16 +1
is divisible by x 2 - x - l.
Solution 21, Alternative 1
Let p and q be the roots of x 2 - x - 1 = 0. By Vieta's theorem,
p + q = 1 and pq = -1. Note that p and q must also be the roots of
ax 17 + bx 16 + 1 = 0. Thus

ap11 + bpl6 = -1 and aq11 + bql6 = -1.

Multiplying the first of these equations by q16 , the second one by p 16 ,
and using the fact that pq = -1, we find

ap + b = -q 16 and aq + b = -p16 . (1)

Thus
3. Solutions to Introductory Problems 41

Since

p+q 1,
p2 +q2 (p+ q) 2 - 2pq = 1+2 = 3,
p4 +q4 (p2 + q2)2 - 2p2q2 = 9 - 2 = 7,
ps + qs (p4 + q4)2 - 2p4q4 = 49 - 2 = 47,

it follows that a= 1 · 3 · 7 · 47 = 987.

Likewise, eliminating a in (1) gives
pl7 - ql7
-b = p-q
= pl6 + pl5q + pl4q2 + ... + ql6
(pl6 + ql6) + pq(pl4 + ql4) + p2q2(pl2 + ql2)
+ ... + p1q1(p2 + q2) + psqs
(pl6 + ql6) - (pl4 + ql4) + ... - (p2 + q2) + l.

For n ~ 1, let k2n = p2n + q2n. Then k2 = 3 and k4 = 7, and

k 2n+4 p2n+4 + q2n+4
(p2n+2 + q2n+2)(p2 + q2) _ p2q2(p2n + q2n)

for n ~ 3. Then ka = 18, ks = 47, k10 = 123, k12 = 322, k14 = 843,
kl6 = 2207.
Hence

-b = 2207 - 843 + 322 - 123 + 47 - 18 + 7 - 3 + 1=1597

or
(a,b) = (987,-1597).

Solution 21, Alternative 2

The other factor is of degree 15 and we write

(c1sx 15 - C14x 14 + · · · + c1x - co)(x 2 - x - 1) = ax 17 + bx 16 + l.

Comparing coefficients:

XO: Co= 1,
X1 : Co - C1 = 0, C1 = 1
x 2 : -ea - C1 + C2 = 0, C2 = 2,
and for 3 ~ k ~ 15, xk : -Ck-2 - Ck-1 +Ck = 0.
42 3. Solutions to Introductory Problems

It follows that fork~ 15, Ck = Fk+l (the Fibonacci number).

Thus a = C1s = F16 = 987 and b = -c14 - c1s = -F11 = -1597 or
(a,b) = (987,-1597).

Comment: Combining the two methods, we obtain some interesting

facts about sequences k2n and F2n-l· Since

it follows that F2n-1 and k2n satisfy the same recursive relation. It is
easy to check that k2 = F1 + F3 and k4 = F3 + Fs.
Therefore k2n = F2n-l + F2n+l and

Problem 22 [AIME 1994]

Given a positive integer n, let p(n) be the product of the non-zero digits
of n. (If n has only one digit, then p(n) is equal to that digit.) Let

s = p(l) + p(2) + ... + p(999).

What is the largest prime factor of S?
Solution 22
Consider each positive integer less than 1000 to be a three-digit number
by prefixing Os to numbers with fewer than three digits. The sum of the
products of the digits of all such positive numbers is

(0 . 0 . 0 + 0 . 0. 1 + ... + 9. 9 . 9) - 0 . 0. 0
= (0 + 1 + ... + 9) 3 - 0.

However, p(n) is the product of non-zero digits of n. The sum of these

products can be found by replacing 0 by 1 in the above expression, since
ignoring O's is equivalent to thinking of them as l's in the products. (Note
that the final 0 in the above expression becomes a 1 and compensates
for the contribution of 000 after it is changed to 111.)
Hence

s = 463 - 1 = (46 - 1)(46 2 + 46 + 1) = 33 . 5. 7. 103,

and the largest prime factor is 103.
3. Solutions to Introductory Problems 43

Problem 23 [Putnam 1979]

Let Xn be a sequence of nonzero real numbers such that
Xn-2Xn-1
Xn=------
2Xn-2 - Xn-1
for n = 3,4, ....
Establish necessary and sufficient conditions on x 1 and x 2 for Xn to be
an integer for infinitely many values of n.
Solution 23, Alternative 1
We have
1 2Xn-2 - Xn-1 2 1
=-----
Xn-2Xn-1 Xn-1 Xn-2

Let Yn = 1/xn. Then Yn - Yn-1 = Yn-1 - Yn-2 7 i.e., Yn is an arithmetic

sequence. If Xn is a nonzero integer when n is in an infinite set S, the
Yn's for n ES satisfy -1 ~ Yn ~ l.
Since an arithmetic sequence is unbounded unless the common difference
is 0, Yn - Yn-1 = 0 for all n, which in turn implies that X1 = X2 = m, a
nonzero integer.
Clearly, this condition is also sufficient.
Solution 23, Alternative 2
An easy induction shows that
X1X2
Xn = ---------
(n - l)x1 - (n - 2)x2
for n = 3,4, ....
In this form we see that Xn will be an integer for infinitely many values
of n if and only if X1 = x2 = m for some nonzero integer m.

Problem 24
Solve the equation
x 3 - 3x = Jx + 2.
Solution 24, Alternative 1
It is clear that x;::: -2. We consider the following cases.
1. -2 ~ x ~ 2. Setting x = 2cosa, 0 ~a~ rr, the equation becomes

8cos3 a - 6cosa = J2(cosa + 1).

or
2cos3a = J4cos 2 ~'
44 3. Solutions to Introductory Problems

from which it follows that cos 3a =cos%·

Then 3a - % = 2m7l', m E Z, or 3a + % = 2n7l', n E Z.
Since 0 :::; a:::; 7!', the solution in this case is

47!' 47!'
x = 2 cos 0 = 2, x = 2 cos 5 , and x = 2 cos 7.

2. x > 2. Then x 3 - 4x = x(x 2 - 4) > 0 and

x2 - x - 2 = (x - 2)(x + 1) > 0

or
x>Jx+2.
It follows that
x3 - 3x > x > J x + 2.
Hence there are no solutions in this case.

Therefore, x = 2, x = 2cos47l'/5, and x = 2cos47l'/7.

Solution 24, Alternative 2
For x > 2, there is a real number t > 1 such that
2 1
x = t + t2.
The equation becomes

3
( t2 + __!__)
t2
- 3 (t2 + __!__)
t2
= Jt2 + __!__
t2
+ 2'

i.e.
6 1 1
t +-=t+-
t6 t'
I.e.
(t 7 - l)(t 5 - 1) = 0,
which has no solutions for t > l.
Hence there are no solutions for x > 2.
For -2 :::; x :::; 2, please see the first solution.
3. Solutions to Introductory Problems 45

Problem 25 [AIME 1992]

For any sequence of real numbers A = {a1, a 2, a 3, · · ·}, define 6.A to be
the sequence {a 2 - a 1, a3 - a2, a4 - a3, ... }.
Suppose that all of the terms of the sequence 6.(6.A) are 1, and that
a19 = a92 = 0.
Find a 1 .
Solution 25
Suppose that the first term of the sequence 6.A is d.
Then
6.A = { d, d + 1, d + 2,. .. }

with the nth term given by d + (n - 1).

Hence

A= {a1,a1 +d,a1 +d+ (d+ l),a1 +d+ (d+ 1) + (d+2),. .. }

with the nth term given by

1
an= a1 + (n - l)d + 2(n - l)(n - 2).

This shows that an is a quadratic polynomial in n with leading coefficient

1/2.
Since a19 = ag 2 = 0, we must have
1
an= 2(n - l9)(n - 92),

so a 1 = (1 - 19)(1 - 92)/2 = 819.

Problem 26 [Korean Mathematics Competition 2000]
Find all real numbers x satisfying the equation

Solution 26
Setting 2x = a and 3x = b, the equation becomes
1 + a2 + b2 - a - b - ab= 0.

Multiplying both sides of the last equation by 2 and completing the

squares gives
(1 - a) 2 +(a - b) 2 + (b - 1) 2 = 0.
46 3. Solutions to Introductory Problems

Therefore 1 = 2x = 3x, and x = 0 is the only solution.

Problem 27 [China 1992]
Prove that
1
16 < z=
80

k=I
~ < 11.
vk

Solution 27
Note that
2(Vk+I-v'k) = ~
k+l+ k
~·
Vk < vk
Therefore
80 1 80
2= Vk > 22= (Vk+1 - v'k) = 16,
k=l k=l

which proves the lower bound.

On the other hand,

2(v'k-Vk=l) = Vkk+ ~
k-1
~·
> vk

Therefore
1
I: Vk < 1+ 2z= (Vk - Vk=l) = 2JSO - 1 < 11,
80 80

k=l k=2

which proves the upper bound. Our proof is complete.

Problem 28 [AHSME 1999]

Determine the number of ordered pairs of integers (m, n) for which mn 2
0 and
m 3 + n 3 + 99mn = 33 3 .

Solution 28
Note that (m + n) 3 = m 3 + n 3 + 3mn(m + n). If m + n = 33, then
33 3 = (m + n) 3 = m 3 + n 3 + 3mn(m + n) = m 3 + n 3 + 99mn.
Hence m +n - 33 is a factor of m 3 + n 3 + 99mn - 33 3 . We have

m3 + n 3 + 99mn - 33 3
= (m + n - 33)(m 2 + n 2 -mn + 33m + 33n + 33 2 )
1
= 2(m + n - 33)[(m - n) 2 + (m + 33) 2 + (n + 33) 2 ].
3. Solutions to Introductory Problems 47

Hence there are 35 solutions altogether: (0, 33), (1, 32), · · ·, (33, 0), and
(-33, -33).

Comment: More generally, we have

a3 + b3 + c3 - 3abc
1
= 2(a + b + c)[(a - b) 2 + (b - c) 2 + (c - a) 2 ].

Problem 29 [Korean Mathematics Competition 2001]

Let a, b, and c be positive real numbers such that a + b + c ::; 4 and
ab + be + ca 2 4.
Prove that at least two of the inequalities

la - bl ::; 2, lb - cl ::; 2, le - al ::; 2

are true.
Solution 29
We have
(a+ b + c) 2 ::; 16,
i.e.
a2 + b2 + c2 + 2(ab +be+ ca) ::; 16,
i.e.
a2 + b2 + c2 ::; 8,
i.e.
a2 + b2 + c2 - (ab+ be+ ca) ::; 4,
i.e.
(a - b) 2 + (b - c) 2 + (c - a) 2 ::; 8,
and the desired result follows.

Problem 30
Evaluate
n 1
2= (n -
k=O
k)!(n + k)! ·
48 3. Solutions to Introductory Problems

Solution 30
Let Sn denote the desired sum. Then

1 n (2n)!
(2n)! ~ (n - k)!(n + k)!
1 2n )
~
n (
(2n)! n-k

~
= 1 n (2n)
(2n)! k

2
(2n)! · 2 ~ k
1 1 [ n (2n) (2n)]
+ n

_1 . ~ [22n + (2n)]
(2n)! 2 n
22n-l 1
= (2n)! + 2(n!) 2 ·

Problem 31 [Romania 1983]

Let 0 < a < l. Solve

for positive numbers x.

Solution 31
Taking loga yields
ax loga x = xa.
Consider functions from JR+ -+ JR,

f(x) =ax, g(x) = loga x, h(x) = xa.

Then both f and g are decreasing and h is increasing. It follows that
f(x)g(x) = h(x) has unique solution x =a.

Problem 32
What is the coefficient of x 2 when

(1+x)(l+2x)(l + 4x) · · · (1+2nx)

is expanded?
3. Solutions to Introductory Problems 49

Solution 32
Let
fn(x) = an,o + an,1X .+ · · · + an,nXn = (1+x)(l+2x) · · · (1+2nx).
It is easy to see that an,o = 1 and
an,l = 1 + 2 + ... + 2n = 2n+l - l.
Since
fn(x) fn-1(x)(l + 2nx)
(1 + (2n - l)x + an-1,2x 2 + · · ·) (1+2nx)
1 + (2n+l - 1) X + (an-1,2 + 22n - 2n) x 2 + · · ·,
we have
an,2 an-1,2 + 22n - 2n
an-2,2 + 22n-2 - 2n-l + 22n - 2n

a1,2 + (24 + 26 + ... + 22n) - (22 + 23 + ... + 2n)

2 + 24(22n-2 - 1) - 4(2n-l - 1)
3
22n+ 2 - 3 · 2n+i + 2 (2n+l - 1) (2n+l - 2)
3 3

Problem 33
Let m and n be distinct positive integers.
Find the maximum value of lxm - xnl, where x is a real number in the
interval (0, 1).
Solution 33
By symmetry, we can assume that m > n. Let y = xm-n.
Since 0 < x < 1, xm < xn and 0 < y < l. Thus
1
lxm - xnl = xn - xm = xn(l - xm-n) = (yn(l - yr-n) rn=n:.
Applying the AM-GM inequality yields

yn(l-yr-n = (m~n)nCm:n)y)n(l-yr-n
< (-n-)n
m-n
(n·~+(m-n)(l-y))n+m-n
n+m-n
50 3. Solutions to Introductory Problems

Therefore

Equality holds if and only if

(m - n)y = l -y
n
or
x= -
m
( n) . . ~n .
Comment: Form= n + 1, we have

xn _ xn+l < ___ nn _

- (n + l)n+l

for real numbers 0 < x < l. Equality holds if and only if x = n/(n + 1).

Problem 34
Prove that the polynomial

where ai, a 2 , ···,an are distinct integers, cannot be written as the prod-
uct of two non-constant polynomials with integer coefficients, i.e., it is
irreducible.
Solution 34
For the sake of contradiction, suppose that

is not irreducible. Let f(x) = p(x)q(x) such that p(x) and q(x) are two
polynomials with integral coefficients having degree less than n. Then

g(x) = p(x) + q(x)

is a polynomial with integral coefficients having degree less than n.
Since
p(ai)q(ai) = f(ai) = -1
and both p(ai) and q(a,) are integers,
3. Solutions to Introductory Problems 51

and
p(ai) + q(ai) = 0.
Thus g(x) has at least n roots. But degg < n, so g(x) = 0. Then
p(x) = -q(x) and f(x) = -p(x) 2 ,
which implies that the leading coefficient of f(x) must be a negative
integer, which is impossible, since the leading coefficient of f(x) is 1.

Problem 35
Find all ordered pairs of real numbers (x, y) for which:
(1 + x)(l + x 2 )(1 + x 4 ) = 1 + y7
and (1 + y)(l + y 2 )(1 + y4 ) = 1 + x7 •

Solution 35
We consider the following cases.
1. xy = 0. Then it is clear that x =y =0 and (x, y) = (0, 0) is a
solution.
2. xy < 0. By the symmetry, we can assume that x > 0 > y. Then
(1 + x)(l + x 2 )(1 + x 4 ) > 1 and 1 + y 7 < 1. There are no solutions
in this case.
3. x, y > 0 and x =I y. By the symmetry, we can assume that x >
y > 0. Then
(1 + x)(l + x 2 )(1 + x 4 ) > 1 + x 7 > 1 + y7 ,
showing that there are no solutions in this case.
4. x, y < 0 and x =I y. By the symmetry, we can assume that x < y <
0. Multiplying by 1- x and 1-y the first and the second equation,
respectively, the system now reads
1 - xB = (1 + y 7 )(1 - x) = 1 - x + y 7 - xy 7
1 - YB = (1 + x 7 )(1 - y) = 1 - y + x 7 - x 7y.
Subtracting the first equation from the second yields
xB - YB = (x - y) + (x1 - y1) - xy(x6 - y6). (1)

Since x < y < 0, xB -yB > 0, x-y < 0, x 7 -y7 < 0, -xy < 0, and
x 6 - y6 > 0. Therefore, the left-hand side of (1) is positive while
the right-hand side of (1) is negative.
Thus there are no solutions in this case.
52 3. Solutions to Introductory Problems

5. x = y. Then solving
1- XS= 1- x + y7 - xy7 = 1 - x + x 7 - XS

leads to x = 0, 1, -1, which implies that (x, y) = (0, 0) or (-1, -1).

Therefore, (x, y) = (0, 0) and (-1, -1) are the only solutions to the
system.

Problem 36
Solve the equation

for real numbers x.

Solution 36
Rearranging terms by powers of 2 yields

2x 2 x + 2x+ 1 (x 2 - 1) - 2(x 2 +x - 1) = 0. (1)

Setting y = x 2 - 1 and dividing by 2 on the both sides, (1) becomes

or
x(2Y - 1) + y(2x - 1) = 0. (2)
Since f(x) = 2x - 1 and x always have the same sign,

x(2Y - 1) · y(2X - 1) ~ 0.

Hence if the terms on the left-hand side of (2) are nonzero, they must
have the same sign, which in turn implies that their sum is not equal to
0.
Therefore (2) is true if and only if x = 0 or y = 0, which leads to solutions
x = -1,0, and l.

Problem 37
Let a be an irrational number and let n be an integer greater than l.
Prove that

is an irrational number.
3. Solutions to Introductory Problems 53

Solution 37
Let
.1. .1.
N = (a + Va2=1")" + (a - Va2=1")" ,
and let
b =(a+ Va2=1r:..
Then N = b + 1/b. For the sake of contradiction, assume that N is
rational. Then by using the identity

repeatedly for m = 1, 2, ... , we obtain that bm + 1/bm is rational for all

mEN.
In particular,

bn +_.!._=a+
bn
Va2=1" +a- Ja2 -1=2a
is rational, in contradiction with the hypothesis.
Therefore our assumption is wrong and N is irrational.

Problem 38
Solve the system of equations

(xi - X2 + X3) 2 X2(X4 + X5 - x2)

(X2 - X3 + X4) 2 = X3(X5 + X1 - X3)

(X3 - X4 + Xs) 2 X4(X1 + X2 - X4)
(x4 - X5 + x1) 2 X5(X2 + X3 - Xs)

(xs - X1 + X2) 2 = X1(X3 + X4 - X1)

Solution 38
Let Xk+s = Xk· Adding the five equations gives
5 5
L(3x~ - 4xkxk+1 + 2xkXk+2) = L(-x~ + 2XkXk+2).
k=l k=l

It follows that
5
L(x~ - XkXk+1) = 0.
k=l
54 3. Solutions to Introductory Problems

Multiplying both sides by 2 and completing the squares yields

5
z)xk - Xk+1) 2 = 0,
k=l

from which x1 = x2 = X3 = X4 = X5. Therefore the solutions to the

system are

for a E JR.

Problem 39
Let x, y, and z be complex numbers such that x+y+z = 2, x2 +y 2 +z 2 =
3, and xyz = 4.
Evaluate
l 1 1
----+
xy + z - 1 yz + x - 1
+----
zx + y - 1

Solution 39
Let S be the desired value. Note that

xy + z - 1 = xy + 1 - x - y = (x - l)(y - 1).

Likewise,
yz +x - 1 = (y - l)(x - 1)
and
zx +y - 1 = (z - l)(x - 1).
Hence
1 1 1
s = (x - l)(y - 1) + (y - l)(z - 1) + -(z---1-)(x---1)
x+y+z-3 -1
(x - l)(y- l)(z - 1) (x - l)(y - l)(z - 1)
-1
xyz - (xy + yz + zx) + x + y + z - 1
-1
5 - (xy + yz + zx)'
But
2(xy + yz + zx) = (x + y + z) 2 - (x 2 + y2 + z2 ) = l.
Therefore S = -2/9.
3. Solutions to Introductory Problems 55

Problem 40 [USSR 1990]

Mr. Fat is going to pick three non-zero real numbers and Mr. Taf is going
to arrange the thr~e numbers as the coefficients of a quadratic equation

_x 2 +_x+ =0.

Mr. Fat wins the game if and only if the resulting equation has two
distinct rational solutions.
Who has a winning strategy?
Solution 40
Mr. Fat has the winning strategy. A set of three distinct rational nonzero
numbers a, b, and c, such that a+ b + c = 0, will do the trick. Let A, B,
and C be any arrangement of a, b, and c, and let f(x) = Ax2 +Bx+ C.
Then
/(1) =A+ B + C =a+ b + c = 0,
which implies that 1 is a solution.
Since the product of the two solutions is C /A, the other solution is C /A,
and it is different from l.

Problem 41 [USAMO 1978]

Given that the real numbers a, b, c, d, and e satisfy simultaneously the
relations

a + b+ c + d + e = 8 and a 2 + b2 + c2 + d2 + e2 = 16,
determine the maximum and the minimum value of a.
Solution 41, Alternative 1
Since the total of b, c, d, and e is 8 - a, their average is x = (8 - a)/4.
Let
b = x +bi, c = x + ci, d = x +di, e = x + ei.
Then bi + ci + di + ei = 0 and
(8 a) 2
16 = a 2 + 4x 2 + b~ + c~ + d~ + e~ 2'. a 2 + 4x 2 = a2 + ~ (1)

or
0 2'. 5a2 - l6a = a(5a - 16).
Therefore 0 ::::; a ::::; 16/5, where a = 0 if and only if b = c =d=e=2
and a = 16/5 if and only if b = c = d = e = 6/5.
56 3. Solutions to Introductory Problems

Solution 41, Alternative 2

By the RMS-AM inequality, (1) follows from

b2 + c2 + d2 + e2 > (b + c + d + e)2 (8 - a) 2
- 4 4
and the rest of the solution is the same.

Problem 42
Find the real zeros of the polynomial

Pa(x) = (x 2 + l)(x - 1) 2 - ax 2 ,
where a is a given real number.
Solution 42
We have
(x 2 + l)(x 2 - 2x + 1) - ax 2 = 0.
Dividing by x 2 yields

(x + ; ) (x - 2+; ) - a = 0.

By setting y = x + l/x, the last equation becomes

y2 - 2y- a= 0.

It follows that
1
x + -x = 1 ± /I+a,
which in turn implies that, if a ~ 0, then the polynomial Pa(x) has the
real zeros
X1,2 =
1 + JI'+'ii ± a + 2JI'+'(i - 2
·
J
2
In addition, if a ~ 8, then Pa(x) also has the real zeros

1- JI'+'ii ± Ja - 2JI'+'(i - 2
X3,4 = 2 ·

Problem 43
Prove that
1 3 2n - 1 1
-·-···--<--
2 4 2n /3ri
for all positive integers n.
3. Solutions to Introductory Problems 57

Solution 43
We prove a stronger statement:
1 3 2n - 1 1
-·-···--<---
2 4 2n - v'3n + (
We use induction.
For n = 1, the result is evident.
Suppose the statement is true for some positive integer k, i.e.,
1 3 2k - 1 1
- . - ... - - < ----,,,==
2 4 2k v'3k+l'
Then
1 3 2k - 1 2k + 1 1 2k + 1
-·- .. · - - · - - < ·--
2 4 2k 2k + 2 v'3k + 1 2k + 2 .
In order for the induction step to pass it suffices to prove that
1 2k + 1 1
---·--<---
v'3k + 1 2k + 2 v'3k + 4.
This reduces to
( 2k +
2k + 2
1) 2
< 3k +
3k +4'
1
i.e.
(4k 2 + 4k + 1)(3k + 4) < (4k 2 + 8k + 4)(3k + 1),
i.e.
0 < k,
which is evident. Our proof is complete.

Comment: By using Stirling numbers, the upper bound can be im-

proved to 1/ ..;;m, for sufficiently large n.

Problem 44 [USAMO Proposal, Gerald Heuer]

Let
P(x) = aoxn + aixn-I +···+an
be a nonzero polynomial with integer coefficients such that

P(r) = P(s) = 0
for some integers rands, with 0 < r < s.
Prove that ak ~ -s for some k.
58 3. Solutions to Introductory Problems

Solution 44
Write P(x) = (x - s)xcQ(x) and

where bm "I- 0. Since Q has a positive root, by Descartes' rule of signs,

either there must exist some k for which bk > 0 ~ bk+l, or bm > 0.
If there exists a k for which bk > 0 ~ bk+l • then

If bm > 0, then am = -sbm ~ -s.

In either case, there is a k such that ak ~ -s, as desired.

Problem 45
Let m be a given real number. Find all complex numbers x such that

Solution 45
Completing the square gives

x
( --1 x )2 2x 2 2
+--1 = -2--1 +m +m,
x+ x- x -
i.e.
2 2 2
( x22x- 1 ) = x22x- 1 + m2 + m.

Setting y = 2x 2/(x 2 - 1), the above equation becomes

y2 - y - (m 2 + m) = 0,

i.e.
(y-m-l)(y+m)=O.
Thus
2x 2 2x 2
- 2- -1 = -m or - 2- -1 = m
x - x -
+ 1,
which leads to solutions

x = ±V m: 2 if m "I- -2 and x = ±J: ~~·if m "I- l.

3. Solutions to Introductory Problems 59

Problem 46
The sequence given by xo = a, x 1 = b, and

Xn+l = ~ (xn-1 + Xln) ·

is periodic.
Prove that ab = l.
Solution 46
Multiplying by 2xn on both sides of the given recursive relation yields

or
2(XnXn+l - 1) = Xn-lXn - l.
Let Yn = Xn-IXn - 1 for n EN. Since Yn+l = Yn/2, {yn} is a geometric
sequence. If Xn is periodic, then so is Yn, which implies that Yn = 0 for
all n EN. Therefore

ab = xox1 = YI +1= l.

Problem 47
Let a, b, c, and d be real numbers such that

(a 2 + b2 - l)(c2 + d2 - 1) > (ac + bd - 1) 2 .

Prove that
a 2 + b2 > 1 and c2 + d2 > l.
Solution 47
For the sake of the contradiction, suppose that one of a 2 + b2 or c2 + d2
is less than or equal to l. Since (ac + bd - 1) 2 ;::: 0, a 2 + b2 - 1 and
c2 + d2 - 1 must have the same sign. Thus both a 2 + b2 and c2 + d 2 are
less than l. Let
x = 1- a2 - b2 and y =1- c2 - d2 .
Then 0 < x, y ~ l. Multiplying by 4 on both sides of the given inequality
gives
4xy > (2ac + 2bd - 2) 2 = (2 - 2ac - 2bd) 2
(a 2+ b2 + x + c2 + d 2 + y - 2ac - 2bd) 2
[(a - c) 2 + (b - d) 2 + x + y] 2
2'.'. (x+y) 2 =x 2 +2xy+y 2 ,
60 3. Solutions to Introductory Problems

or 0 > x 2 2xy + y 2 = (x - y) 2 , which is impossible.

-

Thus our assumption is wrong and both a 2 + b2 and c2 + d2 are greater

than 1.

Problem 48
Find all complex numbers z such that

(3z + 1)(4z + 1)(6z + 1)(12z + 1) = 2.

Solution 48
Note that

8(3z + 1)6(4z + 1)4(6z + 1)2(12z + 1) = 768,

i.e.
+ 8)(24z + 6)(24z + 4)(24z + 2) = 768.
(24z
Setting u = 24z + 5 and w = u 2 yields

(u + 3)(u + l)(u - l)(u - 3) ·= 768,

i.e.
(u 2 - l)(u 2 - 9) = 768,
i.e.
w2 - lOw - 759 = 0,
i.e.
(w - 33)(w + 23) = 0.
Therefore the solutions to the given equation are

z = ±J33 - 5 and z = ±/23i - 5.

24 24

Problem 49
Let X1, X2, · · · , Xn-1, be the zeros different from 1 of the polynomial
P(x) = xn - 1, n ~ 2.
Prove that
1 1 1 n-1
- -+-
1 - X1
- + .. ·+---
1 - X2 1 - Xn-1 2
3. Solutions to Introductory Problems 61

Solution 49, Alternative 1

For i = 1, 2, ... , n, let ai = 1 - Xi. Let
P(l - x) (1 - x)n - 1
Q(x) = = .
x x
Then

and ais are the nonzero roots of the polynomial Q(x), as

Thus the desired sum is the sum of .the reciprocals of the roots of poly-
nomial Q(x), that is,
1 1 1
- -+-
1 - X1
- + .. ·+---
1 - X2 1 - Xn-1
1 1 1
=-+-+
a1 a2
.. ·+--
an-1
a2a3 ···an+ a1a3 ···an+···+ a1a2 · · · an-1
a1a2 ···an

By the Vieta's Theorem, the ratio between

and
P= a1 ···an

is equal to the additive inverse of the ratio between the coefficient of x

and the constant term in Q(x), i.e., the desired value is equal to

s (;) n- l
p - -(~) =-2,

as desired.
Solution 49, Alternative 2
For any polynomial R(x) of degree n-l, whose zeros are x 1, x2, ... , Xn-1'
the following identity holds:
1 1 1 R'(x)
- -+-
x - X1
- + .. ·+---
X - X2 X - Xn-1 R(x).
62 3. Solutions to Introductory Problems

For
R( x ) =Xn - i n-1 n-2 i
--=x +x + ... +x+,
x-i
R(i) = n and

R'(i) = (n - i) + (n - 2) + · · + i = n(n2- i).

It follows that
i i i R'(i) n-i
- - + - - + .. · + - - -
i-x1 i-x2 i-xn-1 R(i) 2

Problem 50
Let a and b be given real numbers.
Solve the system of equations
x-yJx2 -y2
= a,
Ji - x2 + y2

y-xJx2 -y2
b
Ji -x2 +y2
for real numbers x and y.
Solution 50
Let u = x,+ y and v = x - y. Then
u+v u-v
0 < x2 - y 2 = uv < i, x = - 2 -, and y = - 2 -.

Adding the two equations and subtracting the two equations in the orig-
inal system yields the new system
u - u/UV = (a+ b)v'i - uv
v + v/UV = (a - b)v'i - uv.
Multiplying the above two equations yields
uv(i - uv) = (a 2 - b2 )(i - uv),
hence uv = a 2 - b2 . It follows that
(a+ b)v'i - a 2 + b2 (a - b)v'i - a 2 + b2
u = i - v'a2 - b2
and v = -'----=---=-==-
i+v'a2-b2 '
which in turn implies· that

x -(a+bv'a 2 -b2 b+av'a2-b2)

( 'y) - v'i - a 2 +. b2 ' v'i - a2 + b2 '

whenever 0 < a 2 - b2 < 1.

SOLUTlONS TO
ADVANCED PROBLEMS
 4. SOLUTIONS TO
ADVANCED PROBLEMS

Problem 51
Evaluate

( 2000)
2 +
(2000)
5 +
(2000) . . . (2000)
8 + +2000·

Solution 51
Let
f(x) = (1 + x)2ooo = 2000
~
(2000)
k xk.

Let w = (-1 + ../3i)/2. Then w 3 = 1 and w2 +w+1=0. Hence

3 ( (20200) + cooo) + ... + G~~~))

5

= /(1) + wf(w) + w2 f(w 2 )

= 22000 + w(l + w)2000 + w2(l + w2)2000
= 22000 + w(-w2)2000 + w2(-w)2000
= 22000 + w2 + w = 22000 _ 1.

Thus the desired value is

22000 - 1
3

Problem 52
Let x, y, z be positive real numbers such that x 4 + y4 + z4 = 1.
Determine with proof the minimum value of
x3 y3 z3
--+--+--
1 - xS 1 - yS 1 - zS .

Solution 52
For 0 < u < 1, let f(u) = u(l - us). Let A be a positive real number.
By the AM-GM inequality,

A(f(u))s = Aus(l - us) ... (1 - us) ::; [Aus+ 8~1 - us)] 9

66 4. Solutions to Advanced Problems

Setting A= 8 in the above inequality yields

8(f(u)) 8 $ (~) 9

or
8
f(u) S 139 .
It follows that
x3 y3 z3 x4 y4 z4
- - +1-- -y 8+1-
1 - x8
-
- z8
- - - + y(l -
x(l - x 8 ) y8 )
+ ----,---
z(l - z 8 )
(x4 + y4+z4)139
>
8
9{/3
-8-,

with equality if and only if

1
x =y =z = {13"

Comment: This is a simple application of the result of problem 33 in

the previous chapter.

Problem 53 [Romania 1990]

Find all real solutions to the equation

Solution 53
For x < 0, the function f(x) = 2x + 3x + 6x - x 2 is increasing, so the
equation f(x) = 0 has the unique solution x = -1.
Assume that there is a solutions 2'.: 0. Then

s 2 = 25 +3 +65 5 2'.: 3,

sos 2'.: J3, and hence lsJ 2'.: l.

But then s 2'.: ls J yields

25 2'.: 2lsJ = (1 + l)lsJ 2'.: 1 + lsJ 2'.: s,

which in turn implies that

6s > 4s = ( 2s)2 2'.: 8 2.
4. Solutions to Advanced Problems 67

So 28 + 38 + 68 > s 2 , a contradiction.
Therefore x = -1 is the only solution to the equation.

Problem 54
Let {an}n>l be a sequence such that ai = 2 and

for all n EN.

Find an explicit formula for an.
Solution 54
Solving the equation
x 1
x=-+-
2 x
leads to x = ±J2. Note that

an+1+J2 = a~+2J2an+2 = (an+J2)

2

an+l - J2 a; - 2J2an + 2 an - J2
Therefore,

and

Problem 55
Let x, y, and z be positive real numbers. Prove that
x + y
x+ J(x+y)(x+z) y+ J(y+z)(y+x)

z
+ :5 1.
z + 'if (z + x)(z + y)
Solution 55
Note that
J(x + y)(x + z) ~ ..fXY + ..[XZ.
68 4. Solutions to Advanced Problems

In fact, squaring both sides of the above inequality yields

x2 + yz 2'. 2x/JjZ,
which is evident by the AM-GM inequality. Thus

x x ..jX
-x-+-v-;.(;=x=+=y:::;::)=(x=+=z::::;:) ~ x + JXY + y'XZ = _..jX_x_+_JY_y_+_vz_z ·

Likewise,
Y < JY
y + J(y + z)(y + x) - ..jX + JY + vz'
and
z vz
z + J(z + x)(z + y) ~ ../X + JY + .;z·
Adding the last three inequalities leads to the desired result.

Problem 56
Find, with proof, all nonzero polynomials /(z) such that

/(z 2 ) + J(z)f(z + 1) = 0.

Solution 56
Let f(z) = azm(z - 1rg(z), where m and n are non-negative integers
and
g(z) = (z - z1)(z - z2) · · · (z - zk),
Zi # 0 and Zi # 1, for i = 1, 2, ... , k. The given condition becomes

az 2m(z - l)n(z + lt(z 2 - z1)(z 2 - z2) · · · (z 2 - Zk)

= -a 2zm+n(z + 1r(z - l)n(z - z1)(z - z2) · · · (z - zk)
·(z + 1 - z1)(z + 1 - z2) · · · (z + 1 - Zk)·

Thus a = -a 2 , and f is nonzero, so a = -1. Since Zi # 1, 1 - z, # 0.

Then z 2m = zm+n, that is, m = n.
Thus f is of the form
-zm(z - l)mg(z).
Dividing by z 2m(z - l)n(z + l)n, the last equation becomes

g(z 2 ) = g(z)g(z + 1).

4. Solutions to Advanced Problems 69

=
We claim that g(z) 1. Suppose not; then clearly g must have at least
one complex root r =/:- 0. Now

g(r 2 ) = g(r)g(r + 1) = 0,
g(r 4 ) 0,
g(r 8 ) = 0,

and so on.
Since g cannot have infinitely many roots, all its roots must have absolute
value 1.
Now,
g((r - 1)2) = g(r - l)g(r) = 0,
so l(r - 1) 2 1=1.
Clearly, if
lrl = l(r - 1) 2 1= 1,
then
1 + J3i 1 - J3i }
r E { 2 ' 2 .

But r 2 is also a root of g, so the same should be true of r 2 :

2 { 1 + J3i 1 - J3i }
r E 2 ' 2 .

This is absurd. Hence, g cannot have any roots, and g(z) =1.
Therefore, the f(z) are all the polynomials of the form -zm(z - 1r for
mEN.

Problem 57
Let f: N--+ N be a function such that f(n+ 1) > f(n) and f(f(n)) = 3n
for all n.
Evaluate /(2001).
Solution 57, Alternative 1
We prove the following lemma.
Lemma For n = 0, 1, 2, ... ,

1. f (3n) = 2 · 3n; and

2. J (2. 3n) = 3n+1.
70 4. Solutions to Advanced Problems

Proof We use induction.

For n = 0, note that /(l) -:f; 1, otherwise 3 = /(/(1)) = /(1) = 1, which
is impossible. Since f : N --+ N, /(1) > l. Since f(n + 1) > f(n),
f is increasing. Thus 1 < /(1) < /(/(1)) = 3 or /(1) = 2. Hence
/(2) = /(/(1)) = 3.
Suppose that for some positive integer n;:::: 1,

Then,

and
/(2. 3n+l) = j (/(3n+l)) = 3n+2'
as desired. This completes the induction. D
There are 3n - 1 integers m such that 3n < m < 2 · 3n and there are
3n - 1 integers m' such that

Since f is an increasing function,

f (3n + m) = 2 · 3n + m,
for 0 :'.S m :'.S 3n. Therefore

f (2 · 3n + m) = f (f (3n + m)) = 3 (3n + m)

for 0 :'.Sm :'.S 3n. Hence

/(2001) =f (2. 36 + 543) = 3 (36 + 543) = 3816.

Solution 57, Alternative 2

For integer n, let n( 3 ) = ai a2 · · · at denote the base 3 representation of
n.
Using similar inductions as in the first solution, we can prove that

Since 2001( 3 ) = 2202010, /(2001)c 3 ) = 12020100 or

/(2001) = 1 . 32 + 2. 34 + 2. 36 + 1 . 3 = 3816. 7
4. Solutions to Advanced Problems 71

Problem 58 [China 1999]

Let F be the set of all polynomials f(x) with integers coefficients such
that f(x) = 1 has at least one integer root.
For each integer k > 1, find mk, the least integer greater than 1 for which
there exists an f E F such that f(x) = mk has exactly k distinct integer
roots.
Solution 58
Suppose that fk E F satisfies the condition that fk(x) = mk has exactly
k distinct integer roots, and let a be an integer such that fk(a) = 1. Let
9k be the polynomial in F such that

9k(x) = fk(x +a)

for all x.
Now 9k(O) = fk(a) = 1, so the constant term of 9k is 1. Now 9k(x) = mk
has exactly k distinct integer roots ri, r 2 , ... , rk, so we can write

where qk(x) is an integer polynomial.

Note that r1r2 · · · Tk divides the constant term of 9k(x) - mk, which
equals 1 - mk.
Since mk > 1, 1 - mk cannot be 0,

Now r 1 , r 2,: · · rk are distinct integers, and none of them is 0, so

hence
mk ~ lk/2J! · fk/21! + 1.
This value of mk is attained by

9k(x) = (-l)(k;')(x - l)(x + l)(x - 2)(x + 2)

· · · (x + (-l)kfk/21) + lk/2J! · fk/21! + 1.
Thus,
fflk = lk/2J!. fk/21! + 1.
72 4. Solutions to Advanced Problems

Problem 59
Let x 1 = 2 and
Xn+I = x; - Xn + 1,
for n:::: 1.
Prove that
1 1 1 1 1
1-----;;-=r<-+-+···+-<1--n·
22 X1 X2 Xn 22

Solution 59
Since x 1 = 2 and
Xn+I - 1= Xn(Xn - 1).
Xn is increasing.
Then Xn - 1 # 0.
Hence
1 1 1 1
Xn+i-1 Xn(Xn-1) Xn-1 Xn

or
1 1 1
- = - - -Xn+I
- -- -1'
which implies that
1 1 1 1
- + - + .. · + - = l - - - -
X1 X2 Xn Xn+I - 1

Thus it suffices to prove that, for n EN,

1 1 1
1 - 22 n-1 < 1 - l < 1 - 22n •
Xn+I -

or
2n-l 2n
2 < Xn+I - 1<2 . (1)
We use induction to prove (1).
For n = 1, x 2 = x~ - x 1 + 1 = 3 and (1) becomes 2 < 3 < 4, which is
true.
Now suppose that (1) is true for some positive integer n = k, i.e.,

(2)

Then for n = k + 1, the lower bound of (1) follows from

4. Solutions to Advanced Problems 73

Since Xk+i is an integer, the lower bound of (2) implies that

from which it follows that

as desired.
This finishes the induction and we are done.

Problem 60 (Iran 1997]

Suppose that f : R+ --+ R+ is a decreasing function such that for all
x,y ER+,

f(x + y) + f(f(x) + f(y)) = f(f(x + f(y)) + f(y + f(x))).

Prove that f(f(x)) = x.
Solution 60
Setting y = x gives

/(2x) + /(2/(x)) = /(2/(x + f(x))).

Replacing x with f(x) yields

/(2/(x)) + /(2/(/(x))) = /(2/(/(x) + f(f(x)))).

Subtracting these two equations gives

/(2/(/(x))) - /(2x) = /(2/(/(x) + f(f(x)))) - /(2/(x + f(x))).

If f(f(x)) > x, the left hand side of this equa.tion is negative, so
f(f(x) + f(f(x)) > f(x + f(x))
and
f(x) + f(f(x)) < x + f(x),
a contradiction. A similar contradiction occurs if f(f(x)) < x.
Thus f(f(x)) = x as desired.

Comment: In the original formulation f was meant to be a continous

function. The solution above shows that this condition is not necessary.
74 4. Solutions to Advanced Problems

Problem 61 [Nordic Contest 1998]

Find all functions f : Q -+ Q such that

f(x + y) + f(x - y) = 2/(x) + 2/(y)

for all x, y E Q.
Solution 61
The only such functions are f(x) = kx 2 for rational k. Any such function
works, since

f(x + y) + f(x - y) = k(x + y) 2 + k(x - y) 2

= kx 2 + 2kxy + ky 2 + kx 2 - 2kxy + ky 2
= 2kx 2 + 2ky 2
= 2/(x) + 2/(y).
Now suppose f is any function satisfying
f(x + y) + f(x - y) = 2/(x) + 2/(y).

Then letting x = y = 0 gives 2/ (0) = 4/ (0), so f (0) = 0.

We will prove by induction that f(nz) = n 2 f(z) for any positive integer
n and any rational number z.
The claim holds for n = 0 and n = 1; let n ~ 2 and suppose the claim
holds for n - 1 and n - 2.
Then letting x = (n - l)z, y = z in the given equation we obtain

f(nz) + f((n - 2)z) = f((n - l)z + z) + f((n - l)z - z)

= 2/((n - l)z) + 2/(z)

so

f(nz) = 2/((n - l)z) + 2/(z) - f((n - 2)z)

= 2(n - 1) 2 f(z) + 2/(z) - (n - 2) 2 f(z)
= (2n 2 - 4n + 2 + 2 - n 2 + 4n - 4)/(z)
= n 2 f(z)

and the claim holds by induction.

Letting x = 0 in the given equation gives

f(y) + f(-y) = 2/(0) + 2/(y) = 2/(y),

so f(-y) = f(y) for all rational y; thus f(nz) = n 2 f(z) for all integers
n.
4. Solutions to Advanced Problems 75

Now let k = /(1); then for any rational number x = p/q,

q 2 f(x) = f(qx) = f(p) = p 2 /(1) = kp 2

so
J(x) = kp2 / q2 = kx 2 .
Thus the functions f(x) = kx 2 , k E Q, are the only solutions.

Problem 62 [Korean Mathematics Competition 2000]

Let~< a< 1.
Prove that the equation

x 3 (x + 1) = (x + a)(2x +a)
has four distinct real solutions and find these solutions in explicit form.
Solution 62
Look at the given equation as a quadratic equation in a:

a2 + 3xa + 2x2 - x3 - x 4 = 0.

The discriminant of this equation is

9x 2 - 8x 2 + 4x3 + 4x4 = (x + 2x2 ) 2 •

Thus
-3x ± (x + 2x 2 )
a= ---2------.

The first choice a = -x + x 2 yields the quadratic equation x 2 - x - a = 0,

whose solutions are
(1 ± v'f+Ta}
X= 2 .

The second choice a = - 2x - x 2 yields the quadratic equation

x 2 +2x+a = 0,

whose solutions are

-l±Jf=a.
The inequalities

-1 - vT=a < -1 + v'T=CL < 1- JI+"4a < 1 + JI+"4a

2 2
show that the four solutions are distinct.
76 4. Solutions to Advanced Problems

Indeed

~
- 1 +v
1-v1+4a
i - a < - -2- -

reduces to
2Jr=a < 3 - v1 + 4a
which is equivalent to

6Vl + 4a < 6 + 8a,

or 3a < 4a 2 , which is evident.

Problem 63 (Tournament of Towns 1997)

Let a, b, and c be positive real numbers such that abc = l.
Prove that
1 1 1
- - - +
a+b+l b+c+l c+a+l-
+ <l.

Solution 63, Alternative 1

Setting x = a + b, y = b + c and z = c + a, the inequality becomes
1 1 1
--+--+--<1
x+l y+l z+l - '
i.e.
1 1 x
--+--<--
y+l z+l - x+l'
i.e.
_y_+_z_+_2_ < _x_
(y + l)(z + 1) - x + 1'
i.e.
xy + xz + 2x + y + z + 2 ~ xyz + xy + xz + x,
i.e.
x + y + z + 2 ~ xyz,
i.e.
2(a + b + c) + 2 ~ (a+ b)(b + c)(c +a),
i.e.
2(a + b + c) ~ a2 b + ab 2 + b2 c+ bc2 + c2 a ,+ ca 2 .
By the AM-GM inequality,

(a 2 b + a 2 c + 1) 2: 3~ = 3a.
4. Solutions to Advanced Problems 77

Likewise,

and
(c 2 a + c2 b + 1) ~ 3c.
Therefore we only need to prove that

2(a + b + c) + 3 :5 3(a + b + c),

i.e.
3 :5 a+ b + c,
which is evident from AM-GM inequality and abc = l.
Solution 63, Alternative 2
Let a= ay, b=by, c = cY· Then a1b1c1 = 1. Note that

which implies that

Therefore,

1 1
a+b+l
1
<

Likewise,

and
1 b1
---< .
c + a + 1 - a1 + b1 + c1
Adding the three inequalities yields the desired result.
78 4. Solutions to Advanced Problems

Problem 64 (AIME 1988)

Find all functions f, defined on the set of ordered pairs of positive inte-
gers, satisfying the following properties:

f(x, x) = x, f(x, y) = f(y, x), (x + y)f(x, y) = yf(x, x + y).

Solution 64
We claim that f(x, y) = lcm(x, y), the least common multiple of x and
y. It is clear that
lcm(x, x) = x
and
lcm(x, y) = lcm(y, x).
Note that
xy
lcm(x, y) = gcd (x,y )
and
gcd (x, y) = gcd (x, x + y),
where gcd (u, v) denotes the greatest common Q.ivisor of u and v. Then

(x + y)lcm(x, y) = (x + y) · xy
gcd (x, y)

x(x + y)
= y·
gcd (x, x + y)

= ylcm(x, x + y).

Now we prove that there is only one function satisfying the conditions of
the problem.
For the sake of contradiction, assume that there is another function
g(x, y) also satisfying the given conditions.
Let S be the set of all pairs of positive integers (x, y) such that f(x, y) -I
g( x, y), and let (m, n) be such a pair with minimal sum m + n. It is clear
that m -In, otherwise

f(m, n) = f(m, m) = m = g(m, m) = g(m, n).

By symmetry (f(x, y) = f(y, x)), we can assume that n - m > 0.

Note that
nf(m,n -m) [m + (n - m)]/(m, n - m)
(n - m)f(m,m + (n - m))
(n - m)f(m, n)
4. Solutions to Advanced Problems 79

or
n-m
f(m, n - m) = - - · f(m, n).
n
Likewise,
n-m
g(m, n - m} = - - · g(m, n).
n
Since f(m, n) =/:- g(m, n), f(m, n - m) =/:- g(m, n - m).
Thus (m,n-m) ES.
But (m, n - m) has a smaller sum m + (n - m) = n, a contradiction.
Therefore our assumption is wrong and f(x, y) = Icm(x, y) is the only
solution.

Problem 65 [Romania 1990]

Consider n complex numbers zk, such that IZk I :::; 1, k = 1, 2, ... , n.
Prove that there exist e 1 , e2 , ... , en E { -1, 1} such that, for any m :::; n,

Solution 65
Call a finite sequence of complex numbers each with absolute value not
exceeding 1 a green sequence.
Call a green sequence {zk}k=I happy if it has a friend sequence {ek}k=I
of ls and - ls, satisfying the condition of the problem.
We will prove by induction on n that all green sequences are happy.
For n = 2, this claim is obviously true.
Suppose this claim is true when n equals some number m. For the case
of n = m + 1, think of the Zk as points in the complex plane.
For each k, let ek be the line through the origin and the point corre-
sponding to Zk· Among the lines £1 , £2 , £3 , some two are within 60° of
each other; suppose they are fa and £13, with the leftover one being £.-y·
The fact that fa and £13 are withjn 60° of each other implies that there
exists some number er; E { -1, 1} such that z' = Za + e13z13 has absolute
value at most 1.
Now the sequence z', z"I, z4, z5 , ... , Zk+I is a k-term green sequence, so,
by the induction hypothesis, it must be happy; let e', e"I, e4, e5 , ... , ek+I
be its friend.
Let ea= 1.
Then the sequence {ei}~~} is the friend of {zi}~~}. Induction is now
complete.
80 4. Solutions to Advanced Problems

Problem 66 [ARML 1997]

Find a triple of rational numbers (a, b, c) such that

\) ?12 - 1= if<!+ Vb+ ifc;.

Solution 66
Let x = \/~?"2~2~--1 and y = ?"2. Then y 3 = 2 and x = ~y - 1. Note that

1 = y3 - 1 = (y - l)(y 2 + y + 1),
and

2 1 3y2 + 3y + 3 y3 + 3y 2 + 3y + 1 (y + 1)3
y +y+ = 3
3 3
which implies that
1 3
x3 = y- 1= y2 +y +1 (y + 1)3
or
x=--.
?13 (1)
y+l
On the other hand,

3 = y3 +1= (y + l)(y 2 - y + 1)

from which it follows that

1 y2 - y+1
(2)
y+l 3

Combining (1) and (2), we obtain

Consequently,
(a,b,c)=
4 21)
(g'-g'g
is a desired triple.
4. Solutions to Advanced Problems 81

Problem 67 [Romania 1984]

Find the minimum of

where x1, x 2, ... , Xn are real numbers in the interval ( t, 1).

Solution 67
Since loga x is a decreasing function of x when 0 < a < 1 and, since
(x - 1/2) 2 ~ 0 implies x 2 ~ x - 1/4, we have

It follows that

log X2 log X3
> 2 ( --+--+···+ log Xn log x1 )
log X1 log X2 log Xn-1
+--
log Xn

~ 2n

by the AM-GM inequality.

Equalities hold if and only if

X1 = X2 = · · · = Xn = 1/2.

Problem 68 [AIME 1984]

Determine x 2 + y 2 + z 2 + w 2 if
x2 y2 z2 w2
22 - 12 + 22 - 32 + 22 - 52 + 22 - 72 = l,
x2 y2 z2 w2
42 - 12 + 42 - 32 + 42 - 52 + 42 - 72 = 1,

x2 y2 z2 w2
62 - 12 + 62 - 32 + 62 - 52 + 62 - 72 = l,
x2 y2 z2 w2
g2 - 12 + g2 - 32 + g2 - 52 + g2 - 72 = 1.
82 4. Solutions to Advanced Problems

Solution 68
The claim that the given system of equations is satisfied by x 2 , y 2 , z 2 ,
and w2 is equivalent to claiming that

(1)

is satisfied by t = 4, 16, 36, and 64.

Multiplying to clear fractions, we find that for all values oft for which it
is defined (i.e., t f= l, 9, 25, and 49), (1) is equivalent to the polynomial
equation
P(t) = 0,
where

P(t) = (t - l)(t - 9)(t - 25)(t - 49)

-x 2 (t - 9)(t - 25)(t - 49) - y 2 (t - l)(t - 25)(t - 49)
-z 2 (t - l)(t - 9)(t - 49) - w 2 (t - l)(t - 9)(t - 25).

Since degP(t) = 4, P(t) = 0 has exactly four zeros t = 4, 16,36, and 64,
i.e.,
P(t) = (t - 4)(t - 16)(t - 36)(t - 64).
Comparing the coefficients of t 3 in the two expressions of P(t) yields

1+9 + 25 + 49 + x 2 + y 2 + z 2 + w2 = 4+ 16 + 36 + 64,

from which it follows that

x2 + Y2 + z2 + w2 = 36.

Problem 69 [Balkan 1997]

Find all functions f : IR _, IR such that

f(xf(x) + f(y)) = (f(x)) 2 +y

for all x, y E IR.
Solution 69
Let f(O) =a. Setting x = 0 in the given condition yields

f(f(y)) = a 2 + y,

for ally E IR.

Since the range of a 2 +y consists of all real numbers, f must be surjective.
4. Solutions to Advanced Problems 83

Thus there exists b E JR such that f(b) = 0.

Setting x = b in the given condition yields

f(f(y)) = f(bf(b) + f(y)) = (f(b)) 2 +y = y,

for all y E JR. It follows that, for all x, y E JR,

(f(x)) 2 + Y = f(xf(x) + f(y))

= J[f(f(x))f(x) + f(y)] = f[f(x)f(f(x)) + y]
= f(f(x))2 + y = x2 + y,
that is,
(f(x))2 = x2. (1)
It is clear that f(x) = x is a function satisfying the given condition.
Suppose that f(x) "I x. Then there exists some nonzero real number c
such that /(c) = -c. Setting x = cf(c) + f(y) in (1) yields

[f(cf(c) + /(y))] 2 = [c/(c) + /(y)] 2 = [-c2 + /(y)] 2,

for all y E JR, and, setting x = c in the given condition yields

f(cf(c) + f(y)) = (f(c)) 2 + y = c2 + y,

for all y E JR.

Note that (f(y)) 2 = y2.
It follows that
[-c2 + /(y)]2 = (c2 + y)2,
or
f(y) = -y,
for all y E JR, a function which satisfies the given condition.
Therefore the only functions to satisfy the given condition are f(x) = x
or f(x) = -x, for x E JR.

Problem 70
The numbers 1000, 1001, · · ·, 2999 have been written on a board.
Each time, one is allowed to erase two numbers, say, a and b, and replace
them by the number ~min( a, b).
After 1999 such operations, one obtains exactly one number c on the
board.
Prove that c < 1.
84 4. Solutions to Advanced Problems

Solution 70
By symmetry, we may assume a :::; b. Then

~min( a, b) = i·
We have
1 1 1
~+b:::;(i)'
from which it follows that the sum of the reciprocals of all the numbers
on the board is nondecreasing (i.e., the sum is a monovariant).
At the beginning this sum is

where 1/c is the sum at the end. Note that, for 1 :::; k :::; 999,

1 1 4000 4000 1
2000 - k + 2000 + k = 2000 2 - k 2 > 2000 2 = 1000.

Rearranging terms in S yields

or c < 1, as desired.

Problem 71 [Bulgaria 1998)

Let a 1 , a 2 , ... , an be real numbers, not all zero.
Prove that the equation

has at most one nonzero real root.

Solution 71
Notice that fi(x) = y'l + a,x is concave. Hence
4. Solutions to Advanced Problems 85

is concave.
Since f'(x) exists, there can be at most one point on the curve y = l(x)
with derivative 0.
Suppose there is more than one nonzero root.
Since x = 0 is also a root, we have three real roots X1 < X2 < X3. Ap-
plying the Mean-Value theorem to l(x) on intervals [x 1 , x2 ] and [x 2 , x3 ],
we can find two distinct points on the curve with derivative 0, a contra-
diction.
Therefore, our assumption is wrong and there can be at most one nonzero
real root for the equation l(x) = n.

Problem 72 [Turkey 1998]

Let {an} be the sequence of real numbers defined by a 1 = t and
an+l = 4an(l - an)

for n 2'.: 1.
For how many distinct values of t do we have a199S = O?
Solution 72, Alternative 1
Let l(x) = 4x(l - x). Observe that

l- 1 (0) = {O, 1}, l- 1 (1) = {1/2}, l- 1 ([0, 1]) = [O, 1],

and l{Y: l(y) = x}I = 2 for all x E [O, 1).
Let An= {x E JR: r(x) = O}; then

An+1 {x JR:
E r+ (x) = O}
1

{x E JR : r (f (x)) = 0} = { x E JR : I (x) E An}.

We claim that for all n 2'. 1, An C [O, 1], 1 E An, and

IAnl = 2n-l + l.
For n = 1, we have
Ai = {x E JR I l(x) = O} = {O, 1},

and the claims hold.

Now suppose n 2'. 1 and An C [O, 1], 1 E An, and IAnl = 2n-l + 1. Then
x E An+1 ::::} l(x) E An C [O, 1] ::::} x E [O, 1],

so An+i c [O, l].

86 4. Solutions to Advanced Problems

Since /(0) = /(1) = 0, we have r+i(l) = 0 for all n ~ 1, so 1 E An+l·

Now we have

IAn+1 I l{x: f(x) E An}I

= L l{x: f(x) = a}I

l{x: f(x) = l}I + L l{x: f(x) = a}I

aEAn
aE[O,l)

1+ I: 2
aEAn
aE[O,l)

1+2(1Anl - 1)
1 + 2(2n-l + 1 - 1)
2n + 1.

Thus the claim holds by induction.

Finally, a199s = 0 if and only if /1 997 (t) = 0, so there are 2 1996 + 1 such
values oft.
Solution 72, Alternative 2
As in the previous solution, observe that if f(x) E [O, 1] then x E [O, 1],
so if a199s = 0 we must have t E [O, 1].
Now choose 0 E [O, 7r /2] such that sin 0 = /f,.
Observe that for any </J E JR,

f (sin 2 </J) = 4 sin 2 </J (1 - sin 2 </J) = 4 sin 2 </J cos2 </J = sin 2 2¢;
since a 1 = sin 2 0, it foliows that

a2 = sin 2 20, a3 = sin 2 40, ... , a 1998 = sin2 2 1997 0.

Therefore
k7r
a199s =0 {:::::::? sin 21997 0 = 0 {:::::::? 0= 21997

for some k E Z.
Thus the values oft which give a 1998 = 0 are

k E Z, giving 2 1996 + 1 such values of t.

4. Solutions to Advanced Problems 87

Problem 73 [IMO 1997 short list]

(a) Do there exist functions f : JR -+ JR and g : JR -+ JR such that
f(g(x)) = x 2 and g(f(x)) = x 3
for all x E JR?
(b) Do there exist functions f : JR -+ JR and g : JR -+ JR such that
f(g(x)) = x 2 and g(f(x)) = x 4
for all x E JR?

Solution 73
(a) The conditions imply that f(x 3 ) = f(g(f(x))) = [f(x)]2, whence
x E {-1, 0, 1} ===} x3 = x ===} f(x) = [f(x)] 2 ===} f(x) E {O, l}.
Thus, there exist different a, b E {-1, 0, 1} such that f(a) = f(b).
But then = g(f(a)) = g(f(b)) = a contradiction.
a3 b3 ,
Therefore, the desired functions f and g do not exist.
(b) Let
lxlln lxl if lxl 2: 1
g(x)= { lxl-lnlxl ifO<lxl<l
0 if x = 0.
Note that g is even and lal = lbl whenever g(a) = g(b); thus, we
are allowed to define f as an even function such that
f(x) = y2, where y is such that g(±y) = x.
We claim that the functions f, g described above satisfy the condi-
tions of the problem.
It is clear from the definition off that f(g(x)) = x 2.
Now let y = yT[X).
Then g(y) = x and
g(f(x)) = g(y2)
(y2)ln(y 2 ) = y4lny = (ylny)4 if y 2: 1
{ (y2)-ln(y 2 ) = (y-lny)4 ifO<y<l
0 if y = 0
= [g(y)]4
88 4. Solutions to Advanced Problems

Problem 74 [Weichao Wu]

Let 0 < a 1 :::; a2 · · · :::; an, 0 <bi :::; b2 · · · :::; bn be real numbers such that
n n

L::a, 2 I::b;.
i=l i=l

Suppose that there exists 1 :::; k :::; n such that bi :::; a, for 1 :::; i :::; k and
b; 2 a, for i > k.
Prove that

Solution 7 4, Alternative 1
We define two new sequences. For i = 1, 2, ... , n, let

Then
b;ak
1
a· - b
1
= ak - -- = -ak (ai - b,
)
' ' a, a;
or

Therefore

nak = a~ + a; + · · · + a~ 2 b~ + b; + · · · + b~.
Applying the AM-GM inequality yields

from which the desired result follows.

Solution 74, Alternative 2
We define two new sequences. For i = 1, 2, ... , n, let

Then
+ b; + · .. + b~
b~ .::; nak. (1)
Note that, for cy(x - y)(y + c) 2 0,
x x +c
- 2
y
-+
y c
,x2 y and c 2 O;
4. Solutions to Advanced Problems 89

Setting x = ai, y =bi, and c = ak - ai, the above inequality implies that
ai/bi ~ aUbi, for i = 1, 2, ... , n. Thus,

(2)

Using (1) and the AM-GM inequality yields

( alI a2I ... anI ) .ln = ak ~

b~ + b~ + • • • + b~ ~
(b'1b'2 . . . b'n ) .1.n
n
or
a~ a~··· a~ ~ b~ b~ · · · b~. (3)
It is clear that the desired result follows from (2) and (3).

Problem 75
Given eight non-zero real numbers a 1 , a2, ···,as, prove that at least one
of the following six numbers: aia3 + a2a4, aias + a2a6, aia7 + a2as,
a3as + a4a6, a3a7 + a4as, asa7 + a6as is non-negative.
Solution 75 [Moscow Olympiad 1978]
First, we introduce some basic knowledge of vector operations.
Let u = [a, b] and v = [m, n] be two vectors.
Define their dot product u · v = am + bn.
It is easy to check that

(i) v·v = m 2 +n2 = lvl 2, that is, the dot product of vector with itself
is the square of the magnitude of v and v · v ~ 0 with equality if
and only if v = [O, OJ;

(ii) u · v = v · u;
(iii) u · (v + w) = u · v + u · w, where w is a vector;

(iv) (cu)· v = c(u · v), where c is a scalar.

When vectors u and v are placed tail-by-tail at the origin 0, let A and
B be the tips of u and v, respectively. Then AB = v - u.
Let LAOB = 0.
Applying the law of cosines to triangle AO B yields

lv-ul2 = AB2
= OA 2 + OB 2 - 20A · OBcosO
lul 2 + lvl 2 - 2lullvl cosO.
90 4. Solutions to Advanced Problems

It follows that

(v - u) · (v - u) = u · u +v ·v - 2lullvl cosO,

or U·V
cosO= lullvl'
Consequently, if 0 :5 0 :5 90°, u · v 2". 0.
Consider vectors v1 = [a1,a2],v2 = [a3,a4],v3 = [a5,a5], and v4 =
[a1, as].
Note that the numbers a1a3 +a2a4, a1a5 +a2a5, a1a1 +a2as, a3a5 +a4a5,
a3a7 + a4as, a5a7 + a5as are all the dot products of distinct vectors v1
and VJ·
Since there are four vectors, when placed tail-by-tail at the origin, at
least two of them form a non-obtuse angle, which in turn implies the
desired result.

Problem 76 [IMO 1996 short list]

Let a, band c be positive real numbers such that abc = l.
Prove that
~ k ro 1
~+~+~+~+~+k+~+~+ro:5 ·
Solution 76
We have

because
(a3 - b3)(a2 - b2) 2". 0,
with equality if and only if a = b. Hence
ab ab
a5 + b5 +ab :5 a2b2(a + b) +ab
1
ab( a+ b) + 1
abc
= ab(a+b+c)
c
a+b+c·
Likewise,
be < a
b5 + c5 + be - a + b+ c
4. Solutions to Advanced Problems 91

and
ca b
c5 + a 5 +ca ::; a+ b + c ·
Adding the last three inequalities leads to the desired result.
Equality holds if and only if a = b = c = l.
Comment: Please compare the solution to this problem with the
second solution of problem 13 in this chapter.

Problem 77 [Czech-Slovak match 1997]

Find all functions f : JR. ___. JR. such that the equality

f(f(x) + y) = f(x 2 - y) + 4f(x)y

holds for all pairs of real numbers (x, y).
Solution 77
Clearly, f(x) = x 2 satisfies the functional equation.
Now assume that there is a nonzero value a such that f(a) f=. a2.

Let y = x 2 -/(x) in the functional equation to find that

f ( /(x)2+ x2) = f ( /(x)2+ x2) + 2f(x)(x2 - f(x))

or 0 = 2f(x)(x 2 - f(x)). Thus, for each x, either f(x) = 0 or f(x) = x2 .

In both cases, f (0) = 0.
Setting x =a, it follows from above that either f(a) = 0 or f(a) = 0 or
f(a) = a 2 .
The latter is false, so f(a) = 0.
Now, let x = 0 and then x =a in the functional equation to find that

f(y) = f(-y), f(y) = f(a 2 - y)

and so
f(y) = f(-y) = f(a 2 + y);
that is, the function is periodic with nonzero period a 2 .
Let y = a 2 in the original functional equation to obtain

f(f(x)) = f(f(x) + a 2) = f(x 2 - a 2) + 4a 2f(x) = f(x 2) + 4a2f(x).

However, putting y = 0 in the functional equation gives f(f(x)) = f(x 2)
for all x.
92 4. Solutions to Advanced Problems

Thus, 4a 2 f(x) = 0 for all x. Since a is nonzero, f(x) = 0 for all x.

Therefore, either f(x) = x2 or f(x) = 0.

Problem 78 [Kvant]
Solve the system of equations:
3x-y
x+ 2 2 =3
x +y
_ x+3y _ 0
Y x2 + y2 - ·

Solution 78, Alternative 1

Multiplying the second equation by i and adding it to the first equation
yields
. (3x - y) - (x + 3y)i _ 3
x + yi + x 2 +y 2 - '

or
3(x - yi) _ i(x - yi) _ 3
x + yi + 2 2 2 2 - .
x +y x +y
Let z = x + yi. Then
1 x -yi
z x2 + y2.
Thus the last equation becomes

3-i
z+-- =3,
z
or
z2 - 3z + (3 - i) = 0.
Hence
3±
z=-----
J-3 +4i
3±(1+2i)
2 2
that is, (x,y) = (2, 1) or (x,y) = (1, -1).
Solution 78, Alternative 2
Multiplying the first equation by y, the second by x, and adding up yields

2 (3x - y)y - (x + 3y)x _ 3

xy + x2 + y2 - y'

or 2xy - 1 = 3y. It follows that y -1- 0 and

3y + 1
x=~·
4. Solutions to Advanced Problems 93

Substituting this into the second equation of the given system gives

y [ (3"2; I) ' + y'] - (3"2; I) -3y ~ 0,

or
4y 4 - 3y 2 - 1 = 0.

It follows that y 2 = 1 and that the solutions to the system are (2, 1) and
(1, -1).

Problem 79 [China 1995]

Mr. Fat and Mr. Taf play a game with a polynomial of degree at least 4:

x2n + _x2n-1 + _x2n-2 + ... + _x + l.

They fill in real numbers to empty spaces in turn.
If the resulting polynomial has no real root, Mr. Fat wins; otherwise, Mr.
Taf wins.
If Mr. Fat goes first, who has a winning strategy?

Solution 79
Mr. Taf has a winning strategy.
We say a blank space is odd (even) if it is the coefficient of an odd (even)
power of x.
First Mr. Taf will fill in arbitrary real numbers into one of the remaining
even spaces, if there are any.
Since there are only n - 1 even spaces, there will be at least one odd
space left after 2n - 3 plays, that is, the given polynomial becomes

p(x) = q(x) + _xs + _x2t-1,

where s and 2t - 1 are distinct positive integers and q(x) is a fixed

polynomial.
We claim that there is a real number a such that

p(x) = q(x) + axs + _x2t-1

will always have a real root regardless of the coefficient of x 2 t- 1 .
Then Mr. Taf can simply fill in a in front of X 8 and win the game.
94 4. Solutions to Advanced Problems

Now we prove our claim. Let b be the coefficient of x 2t-l in p(x). Note
that
1
22t-l p(2) + p( -1)

= ( 22;-1 q(2) + 2s-2t+1a + b) + [q(-1) + (-l)sa - b]

= ( 22;-1 q(2) + q(-1)) + a[2s-2t+l + (-l)s].

Since s-:/:- 2t-1, 2s- 2t+l + (-1) 5 -:/:- 0.

Thus
1
~q(2) + q(-1)
a= 2s-2t+l + (-l)S
is well defined such that a is independent of b and
1
22t-l p(2) + p(-1) = 0.

It follows that either p( -1) = p(2) = 0 or p( - l) and p(2) have different

signs, which implies that there is a real root of p( x) in between -1 and
2.
In either ca.se, p( x) has a real root regardless of the coefficient of x 2 t- l,
as claimed.
Our proof is thus complete.

Problem 80 [IMO 1997 short list]

Find all positive integers k for which the following statement is true: if
F(x) is a polynomial with integer coefficients satisfying the condition

O~F(c)~k for c=0,1,. . .,k+l,

then F(O) = F(l) = · · · = F(k + 1).

Solution 80
The statement is true if and only if k ;::: 4.
We start by proving that it does hold for each k 2: 4.
Consider any polynomial F(x) with integer coefficients satisfying the
inequality 0 ~ F(c) ~ k for each c E {O, 1, ... , k + 1}.
Note first that F(k + 1) = F(O), since F(k + 1) - F(O) is a multiple of
k + 1 not exceeding k in absolute value.
Hence
F(x) - F(O) = x(x - k - l)G(x),
4. Solutions to Advanced Problems 95

where G(x) is a polynomial with integer coefficients.

Consequently,

k ;:=: IF(c) - F(O)I = c(k + 1 - c)IG(c)I (1)

for each c E {1, 2, ... , k}.

The equality c(k + 1- c) > k holds for each c E {2, 3, ... , k - 1}, as it is
equivalent to (c - l)(k - c) > 0.
Note that the set {2, 3, ... , k - 1} is not empty if k ;:=: 3, and for any c in
this set, (1) implies that IG(c)I < 1.
Since G(c) is an integer, G(c) = 0.
Thus

F(x) - F(O) = x(x - 2)(x - 3) · · · (x - k + l)(x - k - l)H(x), (2)

where H(x) is a polynomial with integer coefficients.

To complete the proof of our claim, it remains to show that H(l) =
H(k) = 0.
Note that for c = 1 and c = k, (2) implies that

k ;:=: IF(c) - F(O)I = (k - 2)! · k · IH(c)I.

For k ;:=: 4, (k - 2)! > 1.

Hence H(c) = 0.
We established that the statement in the question holds for any k ;:=: 4.
But the proof also provides information for the smaller values of k as
well.
More exactly, if F(x) satisfies the given condition then 0 and k + 1 are
roots of F(x) and F(O) for any k ;:=: 1; and if k ;:=: 3 then 2 must also be
a root of F(x) - F(O).
Taking this into account, it is not hard to find the following counterex-
amples:
F(x) = x(2 - x) for k = 1,
F(x) = x(3 - x) for k = 2,
F(x) = x(4 - x)(x - 2) 2 fork= 3.
96 4. Solutions to Advanced Problems

Problem 81 [Korean Mathematics Competition 2001]

The Fibonacci sequence Fn is given by

Prove that
p.
2n -
_ Fln+2 +g Fln-2 _ 2F.3
2n

for all n 2': 2.

Solution 81
Note that

whence
(1)
for all n 2': 2.
Setting a= 3F2n, b = -F2n+2, and c = -F2n-2 in the algebraic identity

a3 + b3 + c3 - 3abc = (a+ b + c)(a2 + b2 + c2 - ab - be - ca)

gives
27 F]n - F]n+2 - Fin-2 - 9F2n+2F2nF2n-2 = 0.
Applying ( 1) twice gives

F2n+2F2n-2 - Fin= (3F2n - F2n-2)F2n-2 - Fin

= F2n(3F2n-2 - F2n) - Fin-2 = F2nF2n-4 - Fin-2
= · · · = F5F2 - Fl = -1.
The desired result follows from
4. Solutions to Advanced Problems 97

Problem 82 [Romania 1998]

Find all functions u : JR - t JR for which there exists a strictly monotonic
function f : JR - t JR such that

f(x + y) = f(x)u(y) + f(y)

for all x, y E JR.
Solution 82
The solutions are u(x) =ax, a E JR+.
To see that these work, take f(x) = x for a= 1.
If a# 1, take f(x) =ax - 1; then

f(x + y) = ax+y - 1 =(ax - l)aY + aY - 1 = f(x)u(y) + f(y)

for all x, y E JR.
Now suppose u : JR - t JR, f : JR - t JR are functions for which f is strictly
monotonic and f(x + y) = f(x)u(y) + f(y) for all x, y E JR.
We must show that u is of the form u(x) =ax for some a E JR+. First,
'letting y = 0, we obtain f(x) = f(x)u(O) + f(O) for all x E JR.
Thus, u(O) # 1 would imply f(x) = /(0)/(1 - u(O)) for all x, which
would contradict the fact that f is strictly monotonic, so we must have
u(O) = 1 and f (0) = 0.
Then f(x) # 0 for all x # 0.
Next, we have

f(x)u(y) + f(y) = f(x + y) = f(x) + f(y)u(x),

or
f(x)(u(y) - 1) = f(y)(u(x) - 1)
for all x, y E JR. That is,
u(x) - 1 u(y) - 1
f (x) f (y)
for all xy # 0.
It follows that there exists CE JR such that
u(x) - 1 = C
f (x)
for all x # 0.
Thus, u(x) = l+Cf(x) for x # O; since u(O) = 1, f(O) = 0, this equation
also holds for x = 0.
98 4. Solutions to Advanced Problems

If C = 0,
then u(x) = 1 for all x, and we are done.
Otherwise, observe

u(x+y) l+Cf(x+y)
1 + Cf(x)u(y) + Cf(y)
u(y) + C f(x)u(y)
= u(x)u(y)

for all x, y E JR.

Thus u(nx) = u(x)n for all n E Z, x E JR.
Since u(x) = 1 + C f(x) for all x, u is strictly monotonic, and u(-x) =
l/u(x) for all x, so u(x) > 0 for all x as u(O) = l.
Let a= u(l) > O; then u(n) =an for all n E N, and

u(p/q) = (u(p)) 1 /q = ap/q

for all p E Z, q E N, so u(x) =ax for all x E Q.

Since u is monotonic and the rationals are dense in JR, we have u(x) =ax
for all x E JR.
Thus all solutions are of the form u( x) = ax, a E JR+.

Problem 83 [China 1986)

Let z1, z2, ... , Zn be complex numbers such that

Prove that there exists a subset S of {zi, z2 , ••• , Zn} such that

Solution 83, Alternative 1

Let £1 , £2 , and £3 be three rays from origin that form angles of 60°, 180°,
and 300°, respectively, with the positive x-axis.
For i = 1, 2, 3, let 'Ri denote the region between ii and ii+l (here f4 = £1 ),
including the ray h Then

By the Pigeonhole Principle, at least one of the above sums is not less
than 1/3.
4. Solutions to Advanced Problems 99

Say it's 'Ra (otherwise, we apply a rotation, which does not effect the
magnitude of a complex number). Let Zk = Xk + iYk· Then for Zk E 'Ra,
Xk = lxkl ;:=: lzkl/2.
Consequently,

as desired.
Solution 83, Alternative 2
We prove a stronger statement: there is subset S of {z1, z2, ... , Zn} such
that

I Lzl;:=:~.
zES

For 1 ~ k ~ n, let Zk = Xk + iYk· Then

1 = lz1l+lz2l+ .. ·+lznl
< (!xii+ IY11) + (Jx2I + IY21) + · · · + (lxnl + IYnl)
L lxkl + L lxkl + L IYkl + L IYkl· Yk<O

By the Pigeonhole Principle, at least one of the above sums is not less
than 1/4. By symmetry, we may assume that

Consequently,

Comment: Using advanced mathematics, the lower bound can be

further improved to l/7r.

Problem 84 [Czech-Slovak Match 1998]

A polynomial P(x) of degree n ~ 5 with integer coefficients and n distinct
integer roots is given.
Find all integer roots of P(P(x)) given that 0 is a root of P(x).
100 4. Solutions to Advanced Problems

Solution 84
The roots of P(x) are clearly integer roots of P(P(x)); we claim there
are no other integer roots.
We prove our claim by contradiction. Suppose, on the contrary, that
P(P(k)) = 0 for some integer k such that P(k)-:/:- 0.
Let
P(x) = a(x - T1)(x - T2)(x - TJ) · · · (x - Tn),
where a, T1, T2, ... , T n are integers,

Since P(k)-:/:- 0, we must have lk - Til 2': 1 for all i.

Since the Ti are all distinct, at most two of lk - T2 I, lk - TJI, lk - T41 equal
1, so

la(k - T2) · · · (k - Tn-1)12': lallk - T2llk - TJllk - T412':2,

and IP(k)I 2': 2lk(k - Tn)I.

Also note that P(k) =Tio for some io, so IP(k)I ::=; ITnl·
Now we consider the following two cases:

1. lkl 2': ITnl· Then IP(k)I 2': 2lk(k - Tn)I 2': 2lkl > ITnl, a contradic-
tion.

2. lkl < ITnl, that is, 1 ::=; lkl ::=; ITnl - 1. Let a, b, c be real numbers,
a ::=; b. For x E [a, b], the function

f(x) = x(c - x)
reaches its minimum value at an endpoint x = a or x = b, or at
both endpoints.
Thus

It follows that

which implies that ITnl ::=; 2. Since n 2': 5, this is only possible if

P(x) = (x + 2)(x + l)x(x - l)(x - 2).

But then it is impossible to have k -:/:-Ti and lkl ::=; ITnl, a contra-
diction.
4. Solutions to Advanced Problems 101

Thus our assumption was incorrect, and the integer roots of P(P(x)) are
exactly all the integer roots of P(x).

Problem 85 [Belarus 1999}

Two real sequences x1, x2, ... , and Y1, Y2, ... , are defined in the following
way:
X1 = Yl = J3, Xn+l = Xn + y'1 + x;;,
and
Yn
Yn+1= l
+ V~ 2
+Yr.
.i

for all n 2:: 1. Prove that 2 < XnYn < 3 for all n > 1.
Solution 85, Alternative 1
Let Zn = 1/Yn and note that the recursion for Yn is equivalent to

Zn+l =Zn+ Vl + z;;.

Also note that z2 = J3 = x1; since the Xis and ZiS satisfy the same
recursion, this means that Zn = Xn-l for all n > 1.
Thus,
Xn Xn
XnYn=-=--.
Zn Xn-1

Note that
J1 + x;,_1 > Xn-1·

Thus Xn > 2Xn-1 and XnYn > 2, which is the lower bound of the desired
inequality.
Since XnS are increasing for n > 1, we have

2 2 1
Xn-1 2:: X1 = 3 > 3'

which implies that

2xn-1 > J1 + x;,_ 1.
Thus 3xn-l > Xn, which leads to the upper bound of the desired inequal-
ity.
Solution 85, Alternative 2
Setting Xn = cot Bn for 0 < Bn < 90° yields

Xn+l =cot Bn + Vl + cot 2 Bn =cot Bn + csc Bn =cot ( (};) .

102 4. Solutions to Advanced Problems

30°
Since (Ji = 30°, we have in general Bn = - -1 . Similar calculation shows
2n-
tha t
2 tan Bn
Yn = tan(2Bn) = 1-tan2 (}n
It follows that
2
xy - - - - -
n n - 1 - tan 2 Bn .
Since tan Bn f:. 0, tan 2 Bn is positive and XnYn > 2.
And since for n > 1 we have Bn < 30°, we also have
2 1
tan Bn <3
so that XnYn < 3.
Comment: From the closed forms for Xn and Yn in the second solution,
we can see the relationship
1
Yn=--
Xn-1

used in the first solution.

Problem 86 [China 1995]

For a polynomial P(x), define the difference of P(x) on the interval [a, b]
([a, b), (a, b), (a, b]) as P(b) - P(a).
Prove that it is possible to dissect the interval [O, 1] into a finite number
of intervals and color them red and blue alternately such that, for every
quadratic polynomial P(x), the total difference of P(x) on red intervals
is equal to that of P(x) on blue intervals.
What about cubic polynomials?
Solution 86
For an interval i, let f:l.ip denote the difference of polynomial P on i.
For a positive real number c and a set S ~ R., let S + c denote the set
obtained by shifting Sin the positive direction by c.
We prove a more general result.
Lemma
Let f. be a positive real number, and let k be a positive integer. It is
always possible to dissect interval Ik = [O, 2kf.] into a finite number of
intervals and color them red and blue alternatively such that, for every
polynomial P(x) with deg P ::; k, the total difference of P(x) on the red
intervals is equal to that on the blue intervals.
4. Solutions to Advanced Problems 103

Proof
We induct on k.
Fork= 1, we can just use intervals [O,t'] and (t',2£]. It is easy to see
that a linear or constant polynomial has the same difference on the two
intervals.
Suppose that the statement is true for k = n, where n is a positive
integer; that is, there exists a set Rn of red disjoint intervals and a set
Bn of blue disjoint intervals such that Rn n Bn = 0, Rn U Bn =In, and,
for any polynomials P(x) with deg P :::;'. n, the total differences of P on
Rn is equal to that of Pon Bn.
Now consider polynomial f(x) with deg f : :;'. n + l. Define

g(x) = f(x + 2nt') and h(x) = f(x) - g(x).

Then deg h :::;'. n. By the induction hypothesis,

I: flbh = I: Llrh,
bEBn rERn

or

It follows that

rER~+l

where

and B~+l

(If R~+I and B~+ 1 both contain the number 2ne, that number may be
removed from one of them.)
It is clear that B~+l and R~+l form a dissection of In+I and, for any
polynomial f with deg f :::;'. n + 1, the total difference off on B~+I is
equal to that off on R~+i ·
The only possible trouble left is that the colors in B~+ 1 UR~+ 1 might not
be alternating (which can happen at the end of the In and the beginning
of In+ 2nt').
But note that if intervals i1 = [a1, b1] and i2 = [b1, c1] are in the same
color, then ·
6.iJ + 6.i2f = 6.i3f,
104 4. Solutions to Advanced Problems

Thus, in B~+l UR~+l' we can simply put consecutive same color intervals
into one bigger interval of the same color.
Thus, there exists a dissection

such that, for every polynomial f(x) with deg f ::; n + 1,

This completes the induction and the proof of the lemma. D

Setting first f. =! and then f. = ~ in the lemma, it is clear that the
answer to each of the given questions is "yes."

Problem 87 (USSR 1990]

Given a cubic equation

x 3 + _x 2 + _x + _ = 0,
Mr. Fat and Mr. Taf are playing the following game.
In one move, Mr. Fat chooses a real number and Mr. Taf puts it in one
of the empty spaces.
After three moves the game is over.
Mr. Fat wins the game if the final equation has three distinct integer
roots.
Who has a winning strategy?
Solution 87
Mr. Fat has a winning strategy.
Let the polynomial be x 3 + ax 2 +bx+ c. Mr. Fat can pick 0 first. We
consider the following cases:

(a) Mr. Taf chooses a= 0, yielding the polynomial equation

x 3 +bx+ c = 0.
Mr. Fat then picks the number -(mnp) 2 , where m, n, and p are
three positive integers such that
m2 +n2 =p2.

If Mr. Taf chooses b = -(mnp) 2 , then Mr. Fat will choose c = 0.

The given polynomial becomes

x(x - mnp)(x + mnp).

4. Solutions to Advanced Problems 105

If Mr. Taf chooses c = -(mnp) 2 , then Mr. Fat will choose

b = m2n2 _ n2p2 _ p2m2.

The given polynomial becomes

(x + m 2 )(x + n2 )(x - p2 ).

(b) Mr. Taf chooses b = 0, yielding the equation

x3 + ax 2 + c = 0.

Mr. Fat then picks the number

m 2(m + 1) 2(m 2 + m + 1) 3,

where m is an integer greater than 1.

If Mr. Taf chooses

a= m 2(m + 1)2(m2 + m + 1) 3,

then Mr. Fat can choose

c = -m 8(m + 1) 8(m 2 + m + 1) 6.

The polynomial becomes

(x - mp)[x + (m + l)p][x + m(m + l)p],

where
p = m 2(m + 1) 2(m 2 + m + 1) 2.

If Mr. Taf chooses

c = m 2(m + 1) 2(m 2 + m + 1)3,

then Mr. Fat can choose

a=-(m 2 +m+1) 2.

The polynomial becomes

(x + mq)[x - (m + l)q][x - m(m + l)q],

where
q = m 2 +m+1.
106 4. Solutions to Advanced Problems

(c) Mr. Taf chooses c = 0.

Then the problem reduces to problem 40 of the previous chapter.
Mr. Fat needs only to pick two integers a and b such that

ab(a - l)(b - 1) =f:. 0

and a+ b = -1.
The polynomial becomes either x(x - l)(x - a) or x(x - l)(x - b).

Our proof is complete.

Below is an example of what Mr. Fat and Mr. Taf could do:

F T F T F Roots
0 a -3600 b 0 -60,0, 60
,, ,, ,, c -481 -16, -9, 25
,, b 4.9.7 a -2 .3 .7 -8. 27. 49,
-4. 27. 49,
8·9·49
,, ,, ,, c -49 -14, 21, 42
,, c 2 a -3 -3,0, 1
,, ,, ,, b -3 0, 1,2

Problem 88 [Romania 1996]

Let n > 2 be an integer and let f : IR2 --+ IR be a function such that for
any regular n-gon A1A2 ... An,

Prove that f is the zero function.

Solution 88
We identify IR 2 with the complex plane and let ( = e27ri/n.
Then the condition is that for any z EC and any positive real t,
n
I:t(z + t(i) = o.
i'=l

In particular, for each of k = 1, ... , n, we obtain

n
I: t (z - (k + (i) = o.
j=l
4. Solutions to Advanced Problems 107

Summing over k, we have

n n
L Lf(z - (1- (m)(k) = 0.
m=l k=l

Form= n the inner sum is nf(z); for other m, the inner sum again runs
over a regular polygon, hence is 0.
Thus f(z) = 0 for all z E C.

Problem 89 [IMO 1997 short list]

Let p be a prime number and let f(x) be a polynomial of degreed with
integer coefficients such that:
(i) /(0) = 0, /(1) = 1;
(ii) for every positive integer n, the remainder upon division of f(n)
by pis either 0 or 1.
Prove that d ~ p - 1.
Solution 89, Alternative 1
For the sake of the contradiction, assume that d :5 p - 2.
Then by Lagrange's interpolation formula the polynomial f(x) is
determined by its values at 0, 1, ... , p - 2; that is,

f(x) = '"°' f(k) x · · · (x - k + l)(x - k - 1) · · · (x - p + 2)

p-2

6 k···l·(-l)···(k-p+2)

'"°'
p-2

6f(k)
x · · · (x - k + 1) (x - k - 1) · · · (x - p + 2)
k!(-l)P-k (p-k-2)! .

Setting x = p - 1 gives
p-2
f(p - 1) = '"'f(k) (p- l)(p - 2) ... (p- k)
~ (-l)P-kk!
k=O

p-2 (-l)kk'
L f(k) (-l)P-kk!
k=O
p-2
- (-l)PLf(k) (mod p).
k=O

It follows that

S(f) := /(0) + /(1) + · · · + f(p - 1) =0 (mod p). (1)

108 4. Solutions to Advanced Problems

On the other hand, (ii) implies that S(f) = J (mod p), where j denotes
the number of those k E {O, 1, ... , p- 1} for which f(k) = 1 (mod p).
But (i) implies that 1 S J S p - 1.
So S(f) =/=. 0 (mod p), which contradicts (1).
Thus our original assumption was wrong, and our proof is complete.
Solution 89, Alternative 2
Again, we approach the problem indirectly.
Assume that d Sp - 2, and let

J(x) = ap-2xp- 2 + · · · + alx + ao.

Then
p-1 p-1 p-2 p-2 p-1 p-2

S(f) = Lf(k) = LLaiki = z=a, Lki = LaiSi,

k=O k=Oi=O i=O k=O i=O

p-1

where Si= L ki.

k=O
We claim that Si= 0 (mod p) for all i = 0, 1, ... , p - 2.
We use strong induction on i to prove our claim.
The statement is true for i = 0 as So = p.
Now suppose that So
p - 2. Note that
= =···= =0
S1 Si-1 (mod p) for some 1 S i S

k=l k=O k=O

=I: t
k=O j=O
(i ~ 1) ki (i + +I: (i ~ 1) Si
J
= l)Si
j=O J
= (i + l)Si (mod p)

Since 0<i+1 < p, it follows that Si= 0 (mod p). This completes the
induction and the proof of the claim. Therefore,
p-2

S(f) = L aiSi =0 (mod p).

i=O

The rest is the same as in the first solution.

4. Solutions to Advanced Problems 109

Problem 90
Let n be a given positive integer.
Consider the sequence ao, ai, · · · , an with ao = ~ and

fork= 1, 2, · · ·, n.
Prove that
1
l--<an<l.
n

Solution 90, Alternative 1

We prove a stronger statement: Fork= 1, 2, ... , n,
n+ 1 n
-2n___k_+_2 < ak < -2n---k · (1)

We use induction to prove both inequalities.

We first prove the upper bound. For k = 1, it is easy to check that
1 1 2n + 1 n
ai = 2 + 4n = ~ < 2n - 1 ·

Suppose that
n
ak < 2n - k'
for some positive integer k < n. Then

< 2n ~ k ( n + 2n ~ k)
n(2n - k + 1)
= (2n - k) 2
n
< 2n - k - 1'
as
(2n - k + 1)(2n - k - 1) = (2n - k) 2 - 1 < (2n - k)2.
Thus our induction step is complete. In particular, for k =n- 1,

n -l
an = ak+l < 2n - (n - 1) - 1 - '
110 4. Solutions to Advanced Problems

as desired.
Now we prove the upper bound. Fork= 1, it is easy to check that
2n + 1 n+1
a 1 =~ > 2n+l·

Suppose that
n+l
ak > 2n - k +2'
for some positive integer k < n. Then

a~ n+1 (n + 1) 2
ak+i = ak + ~ > 2n - k + 2 + n(2n - k + 2) 2
It follows that
n+l > n+l (n+1) 2
ak+l - 2n - k + 1 - (2n - k + 1)(2n - k + 2)
+n(2n
- -- - --
k + 2) 2

n+ 1 (n+ 1 2n - k + 2)
2n - k + 2)2 n 2n - k + 1

= n+1
2n - k + 2)2
(.!.n _2n - 1k + 1) >0
·
This complete the induction step. In particular, for n = k - 1, we obtain
n+l n+l 1 1
an= ak+l > 2n- (n- l) + l = --
n+ 2
= 1- - -
n+ 2
> 1- -,
n
as desired.
Solution 90, Alternative 2
Rewriting the given condition as
1 1 n 1 1
= -----=-- ------
aL1
ak-1 +--
n
yields
1 1 1
(2)
n + ak-1'
for k = 1, 2, ... , n.
It is clear that aks are increasing.
Thus
1
an > an-1 > ... > ao = 2·
4. Solutions to Advanced Problems 111

Thus (2) implies that

1 1 1
----<-
ak-1 ak n
for k = 1, 2, ... , n.
Telescoping summation gives
1 1
---<1
ao an
or
1 1
-
> - - 1 = 2 - 1 = 1,
an ao
that is, an < 1, which gives the desired upper bound.
1
Since an < 1, and, since aks are increasing, "2 = ao < ak ~ an < 1 for
k = 1,2, ... ,n.
Then (2) implies
1 1 1 1
>--1,
n+ak-1 n+

for k = 1, 2, ... , n.
Telescoping sum gives
1 1 n
--->--
ao an n +1
or
1 1 n n+2
-<-------
an ao n+l - n+l'
that is,
n+ 1 1 1
an > -- = 1 - -- > 1- -
n+2 n+2 n'
which is the desired lower bound.

Problem 91 [IMO 1996 short list]

Let a 1 , a2, ... , an be nonnegative real numbers, not all zero.
(a) Prove that xn - a 1xn-l - · · · - an-1X - an = 0 has precisely one
positive real root R.

(b) Let A= 'L'j= 1 a3 and B = 'L'j= 1 ja3.

Prove that AA~ R 8 .
112 4. Solutions to Advanced Problems

Solution 91

(a) Consider the function

Note that f decreases from oo to 0 as x increases from 0 to oo.

Hence there is•a unique real number R such that f(R) = 1, that is,
there exists a unique positive real root R of the given polynomial.

(b) Let Cj = ai/A.

Then CjS are non-negative and L Cj = l.
Since - In x is a convex function on the interval (0, oo), by Jensen's
inequality,

tc
J=l
3 (-In :j) ~-In (tc ~) = -ln(f(R)) =
J=l
3 0.

It follows that
n
L Cj ( - In A +j In R) ~ O
j=l
or n n
L Cj lnA::; L)cj lnR.
j=l j=l

Substituting c3 = aj/A, we obtain the desired inequality.

Comment: Please compare the solution of (a) with that of the problem
15 in the last chapter.

Problem 92
Prove that there exists a polynomial P(x, y) with real coefficients such
that P(x, y) ~ 0 for all real numbers x and y, which cannot be written
as the sum of squares of polynomials with real coefficients.
Solution 92
We claim that
1
P(x, y) = (x2 + Y2 - l)x2y2 + 27
is a polynomial satisfying the given conditions.
First we prove that P(x, y) ~ 0 for all real numbers x and y.
4. Solutions to Advanced Problems 113

If x 2 + y2 - 1 ~ 0, then it is clear that P(x, y) > O; if x 2 + y 2 - 1 < 0,

then applying the AM-GM inequality gives

1
27'
or
1
(x2 + y2 - l)x2y2 ~ - 27·

It follows that P(x, y) ~ 0.

We are left to prove that P(x, y) cannot be written as the sum of squares
of polynomials with real coefficients.
For the sake of contradiction, assume that
n
P(x,y) = LQi(x,y) 2.
i=l

Since deg P = 6, deg Qi :::; 3.

Thus

Aix3 + Bix 2y + Cixy 2 + Diy 3

+Eix2 + Fixy + Giy 2 +Hix+ Iiy +Ji.

Comparing the coefficients, in P(x,y) and E~ 1 Qi(x,y) 2, of terms x 6

and y6 gives
n n
LA~= LD~=O,
i=l i=l

or Ai = Di = 0 for all i.
Then, comparing those of x 4 and y 4 gives
n n
LEI=LG~=O,
i=l i=l

or Ei =Ci= 0 for all i.

Next, comparing those of x 2 and y 2 gives
n n
LHI = LII =0,
i=l i=l

or Hi = Ii = 0 for all i.
Thus,
114 4. Solutions to Advanced Problems

But, finally, comparing the coefficients of the term x 2 y 2 , we have

n
LFl =-1,
i=l

which is impossible for real numbers Fi.

Thus our assumption is wrong, and our proof is complete.

Problem 93 [IMO 1996 short list]

For each positive integer n, show that there exists a positive integer k
such that
k = f(x)(x + 1) 2 n + g(x)(x 2n + 1)

for some polynomials f, g with integer coefficients, and find the smallest
such k as a function of n.
Solution 93
First we show that such a k exists.
Note that x + 1 divides 1 - x 2n. Then for some polynomial a(x) with
integer coefficients, we have

(1 + x)a(x) = 1 - x 2n = 2- (1 + x 2n),
or
2 = (1 + x)a(x) + (1 + x 2n).
Raising both sides to the (2n)th power, we obtain

22n = (1 + x)2n(a(x))2n + (1 + x2n)b(x),

where b(x) is a polynomial with integer coefficients.

This shows that a k satisfying the condition of the problem exists. Let
k0 be the minimum such k.
Let 2n = 2r · q, where r is a positive integer and q is an odd integer.
We claim that ko = 2q.
First we prove that 2q divides k 0 . Let t = 2r. Note that x 2 n +1 =
(xt + l)Q(x), where
Q(x) = xt(q-l) _ xt(q-2) + ... _ xt + l.
The roots of xt + 1 are

Wm = cos ( (2m-1)7l')
t + i..sm ((2m-1)7l')
t , m = 1,2, ... ,t;
4. Solutions to Advanced Problems 115

that is,
R(x) = xt + 1 = (x -wi)(x - w2) · · · (x - Wt)·
Let f(x) and g(x) be polynomials with integer coefficients such that

ko = f(x)(x + 1) 2n + g(x)(x 2n + 1)
f(x)(x + 1) 2n + g(x)Q(x)(xt + 1).

It follows that

(1)

Since r is positive, t is even. So

Since f (w1) f (w2) · · · f (Wt) is a symmetric polynomial in wi, w2, ... , Wt

with integer coefficients, it can be expressed as a polynomial with integer
coefficients in the elementary symmetric functions in w1 , w2 , ... , Wt
and therefore

is an integer.
Taking the product over m = 1, 2, ... , t, (1) gives 22 n F = ki or 22 r·qp =
k'{. It follows that 2q divides ko.
It now suffices to prove that ko ~ 2q.
Note that Q(-1) = 1.
It follows that
Q(x) = (x + l)c(x) + 1,
where c(x) is a polynomial with integer coefficients.
Hence
(x + 1) 2n(c(x)) 2n = (Q(x) - 1) 2n = Q(x)d(x) + 1, (2)
for some polynomial d( x) with integer coefficients.
Also observe that, for any fixed m,
2i-l .
{ Wm . J. -- 1, 2,. .. , t} -- {Wi,W2,. . .,Wt }.

Thus
(1 +wm)(l +w~J · · · (1 +w~- 1 ) = R(-1) = 2,
and writing
116 4. Solutions to Advanced Problems

we find that for some polynomial h(x), independent of m, with integer

coefficients such that
(1 + Wm)th(wm) = 2.
But then (x + l)h(x) - 2 is divisible by xt + 1 and hence we can write

(x + l)h(x) = 2 + (xt + l)u(x),

for some polynomial u(x) with integer coefficients.
Raising both sides to the power q we obtain

(x + l) 2 n(h(x))q = 2q + (xt + l)v(x), (3)

where v(x) is a polynomial with integer coefficients.

Using (2) and (3) we obtain

(x+ 1) 2 n(c(x)) 2 n(xt + l)v(x)

= Q(x)d(x)(xt + l)v(x) + (xt + l)v(x)
= Q(x)d(x)(xt + l)v(x) + (x + l) 2 n(h(x))q - 2q,

that is,
2q = fi(x)(x + 1) 2n + g1(x)(x 2 n + 1),
where Ji (x) and 91 ( x) are polynomials with integer coefficients.
Hence ko ~ 2q, as desired.
Our proof is thus complete.

Problem 94 [USAMO 1998 proposal, Kiran Kedlaya]

Let x be a positive real number.
(a) Prove that
00
(n - 1)! 1
f; (x + 1) · .. (x + n) = ~·

(b) Prove that

00
(n-1)! 00
1
f; n(x + 1) ... (x + n) = ~ (x + k)2 ·
4. Solutions to Advanced Problems 117

Solution 94
We use infinite telescoping sums to solve the problem.

(a) Equivalently, we have to show that

~
00 n.x' _ 1
~ n(x + 1) · · · (x + n) - ·
Note that
x 1 1
n(x+n) n x+n
It follows that
n!x
n(x + 1) · · · (x + n)
(n -1)! n!
(x + 1) · · · (x + n - 1) (x + 1) · · · (x + n)'
and this telescoping summation yields the desired result.

(b) Let
00
(n - 1)!
f(x) = ~ n(x + 1) ... (x + n)"
1
Then, by (a), f(x) < -.
x
In particular, f(x) converges to 0 as x approaches oo, so we can
write f as an infinite telescoping series
00

f(x) = L[f(x + k - 1) - f(x + k)]. (1)

k=l

On the other hand, the result in (a) gives

f(x - 1) - f(x) = ~ (n - 1)! (1 1 )

~ n(x + 1) ... (x + n - 1) ; - x + n

1 00
(n-1)!
= ; ~ (x + 1) · · · (x + n)
1
x2.
118 4. Solutions to Advanced Problems

Substituting the last equation to (1) gives

00 1
f(x) = ~ (x + k)2,
as desired.

Problem 95 [Romania 1996]

Let n 2". 3 be an integer, and let

X ~ S = {1, 2, ... , n 3 }

be a set of 3n 2 elements.
Prove that one can find nine distinct numbers ai, bi, c, (i = 1, 2, 3) in X
such that the system

a1x + b1y + c1z 0

a2x + b2y + c2z 0
a3X + b3y + C3Z Q

has a solution (xo, Yo, zo) in nonzero integers.

Solution 95
Label the elements of X in increasing order x 1 < · · · < x 3 n2, and put
X1 {x1,. .. ,Xn2},
X2 {xn2+1, ... ,X2n2},
X3 {x2n2+1,. . .,X3n2}.

Define the function f: X 1 x X 2 x X3 -+ S x Sas follows:

f(a,b,c) = (b-a,c-b).

The domain off contains n 6 elements.

The range of f, on the other hand, is contained in the subset of S x S
of pairs whose sum is at most n 3 , a set of cardinality
n3
-1 n 3 (n 3 - 1) n6
L: k=
k=l
2 < 2·
By the Pigeonhole Principle, some three triples (ai, bi, c,) (i = 1, 2, 3)
map to the same pair, in which case x ·= b1 - c1, y = c1 - a 1, z = a 1 - b1
is a solution in nonzero integers.
4. Solutions to Advanced Problems 119

Note that ai cannot equal bi since X 1 and X 2 are disjoint, and that
a 1 = a2 implies that the triples (a 1, bi, c1) and (a2, b2, c2) are identical,
a contradiction.
Hence the nine numbers chosen are indeed distinct.

Problem 96 [Xuanguo Huang]

Let n ~ 3 be an integer and let x 1, x2, · · · , Xn be positive real numbers.
Suppose that
n 1
" --1
~l+x· - ·
j=l J

Prove that
1 1 1 )
Fi+./X2+···+..fi;i~(n-l) ( -+-+···+- .
y'x1 .jX2 Fn
Solution 96
By symmetry, we may assume that X1 ~ X2 ~ · · · ~ Xn· We have the
following lemma.
Lemma For 1 ~ i <j ~ n,

.;x;. > vx; .

1 +Xi - 1 + Xj

Proof Since n ~ 3, and, since

n 1
" - = 1'
~l+x·
i=l i

we have

or
1 +Xi+ Xj + XiXj > 2 +Xi+ Xj.
It follows that XiXj > l. Thus
.;x;. - vx; = .jX;.(1 + Xj) - .jX;(l +Xi)
1 +Xi 1 + Xj (1 + Xi)(l + Xj)
(.;x;. - .jXj) (1 - .fliXJ)
=
(1 + Xi)(l + Xj)

~ 0,
120 4. Solutions to Advanced Problems

as desired. D

By the lemma, we have

- > - Vx2
-Fi - > · · · >Fn
--
1 + X1 - 1 + X2 - - 1 + Xn '

and, since
1 1 1
-->-->···>--
Fi - Vx2 - - Fn'
it follows by the Chebyshev Inequality

(1)

By the Cauchy-Schwartz Inequality, we have

or
(2)

Multiplying by
1 n 1
~I: tx:
i=l v ""'•

on both sides of (2) and applying (1) gives

which in turn implies the desired inequality.

4. Solutions to Advanced Problems 121

Problem 97
Let XI, x2, ... , Xn be distinct real numbers.
Define the polynomials

P(x) = (x - xI)(x - x2) · · · (x - Xn)

and
Q(x) = P(x) ( -1- +- 1- + · .. + -1- ) .
X - XI X - X2 X - Xn
Let YI, y2, ... , Yn-I be the roots of Q.
Show that
min
ioj.j lxi - x J I< min
i#j IYi -y·I·
J

Solution 97
By symmetry, we may assume that

d = rajn
i.,-J
IYi - Yi I = Y2 - YI·

Let Sk =YI - xk, fork= 1, 2, ... , n.

By symmetry, we may also assume that sI < s2 < · · · < sn, i.e., XI >
X2 > · · · > Xn·

For the sake of contradiction, assume that

d:::; min lxi - xii= min xi - Xj = mins 1 - si. (1)

i#J i<J i<J

Since P has no double roots, it shares none with Q.

Then

or
1 1 1
- - + Yi -
Yi - XI X2
+···+ Yi - Xn
=0.

In particular, setting i = 1 and i = 2 gives

n 1 n 1
L Sk = k=I
k=I
L Sk + d = O. (2)

We claim that there is a k such that sk(sk + d) < 0, otherwise, we have

1 1
--<-
Sk +d Sk
122 4. Solutions to Advanced Problems

for all k, which in turn implies that

n 1 n 1
Lsk+d <Lsk'
k=l k=l

which contradicts (2).

Let j be the number of ks such that sk(sk+d) < 0, that is, Sk < 0 < sk+d.
A simple but critical fact is that Sk + d and Sk+j have the same sign. In
fact, suppose that

S1 < · · · < Si < Si+l < · · · < Si+j < 0 < Si+j+l < · · · < Sni
then
+ d < · · · < Si + d < 0 < Si+l + d < · · · < Sn + d.
S1

Then Sk+j > 0 if and only if k > i + 1, that is Sk + d > 0.

From (1), we obtain sk + d :£ Sk+j, and, since Sk + d and Sk+j have the
same sign, we obtain
1 1
-->--
Sk +d - Sk+j'

for all k = 1, 2, · · ·, n - j. Therefore,

n-j 1 n-j 1
'"'-<'"'-
k=l
+
L.,, Sk+ · - L.,, Sk
J k=l
d'

or
n 1 n-j 1
2.:-<l.:-· (3)
k=j+l Sk - k=l Sk +d
Also note that
j 1 n 1
2.:-<o< l.: --. (4)
k=l Sk k=n-j+l Sk +d
Adding (3) and (4) yields
n 1 n 1
Lsk < Lsk+d'
k=l k=l

which contradicts (2).

Thus our assumption is wrong and our proof is compJete.
4. Solutions to Advanced Problems 123

Problem 98 [Romania 1998]

Show that for any positive integer n, the polynomial

f(x) = (x 2 + x)2'' +1
cannot be written as the product of two non-constant polynomials with
integer coefficients.
Solution 98
Note that f(x) = g(h(x)), where h(x) = x 2 + x and g(y) = y2n + l.
Since

and (2kn) is even for 1 ~ k ~ 2n - 1, g is irreducible, by Eisenstein's

criterion.
Now let p be a non-constant factor of f, and let r be a root of p.
Then g(h(r)) = f(r) = 0, sos:= h(r) is a root of g.
Since s = r 2 + r E Q(r), we have Q(s) C Q(r), so

deg p ~ deg(Q(r)/Q) ~ deg(Q(s)/Q) =deg g = 2n.

Thus every factor of f has degree at least 2n.

Therefore, if f is reducible, we can write f(x) = A(x)B(x) where A and
B have degree 2n.
Next, observe that

f(x) = (x2 + x)2n + 1

x 2 n+l + x 2 + 1 =: (x 2 + X + 1) 2
n n (mod 2).

Since x 2 + x + 1 is irreducible in Z 2 [x], by unique factorization we must

have

A(x) =B(x) =(x 2 + x + 1) =x 2 + x 2 n-i n 2 n-l +1 (mod 2).

Thus, if we write
2n
A(x) a2nX + · · · + ao,
B(x) b2nX 2n + · · · + bo,

then a2n, a 2 n-1, ao, b2n, b2 n-1, bo are odd and all the other coefficients
are even. Since f is monic, we may assume without loss of generality
124 4. Solutions to Advanced Problems

that a2n = b2n = 1; also, aobo = f (0) = 1, but ao > 0, bo > 0 as f has
no real roots, so ao = bo = 1.
Therefore,

as a 2 n-1 + b2 n-1 is even.

But

and ( 2 ~~~~ 1 ) is odd by Lucas's theorem, so

a contradiction.
Hence f is irreducible.

Problem 99 [Iran 1998]

Let Ji, h, /J : JR --+ JR be functions such that

is monotonic for all a 1, a2, a3 E JR.

Prove that there exist c1 , c2 , c3 E JR, not all zero, such that

for all x E JR.

Solution 99, Alternative 1
We establish the following lemma.
Lemma: Let f, g : JR --+ JR be functions such that f is nonconstant and
af + bg is monotonic for all a, b E JR. Then there exists c E JR such that
g - cf is a constant function.
Proof Lets, t be two real numbers such that /(s) =J. f(t).
4. Solutions to Advanced Problems 125

Let
g(s) - g(t)
u - ---'-"'-------'--'--
- f(s) - f(t) ·
Let hi = g - di/ for some di E JR.
Then hi is monotonic. But

hi(s) - hi(t) = g(s) - g(t) - di(f(s) - f(t)) = (f(s) - f(t))(u - di).

Since f(s) - f(t) # 0 is fixed, the monotonicity of hi depends only on

the sign of u - di.
Since f is nonconstant, there exist xi, X2 E JR such that f (xi) # f (x2).
Let

and h = g - cf.
Then r = h(xi) = h(x2) and the monotonicity of hi = g - dif, for each
di, depends only on the sign of c - di.
We claim that h = g - cf is a constant function.
We prove our claim by contradiction.
Suppose, on the contrary, that there exists x 3 E JR such that h(x3 ) # r.
Since /(xi) # f(x2), at least one of f(x1) # f(xa) and f(x2) # f(xa) is
true.
Without loss of generality, suppose that /(xi) # f(x 3 ).
Let
, g(xi) - g(xa)
c = .
/(xi) - f(xa)
Then the monotonicity of hi also depends only on the sign of c' - di.
Since h(xa) # r = h(xi),

c # g(xi) - g(xa) _ c'·

/(xi) - f(xa) - '

hence c - di # c' - di.

So there exists some di such that hi is both strictly increasing and de-
creasing, which is impossible.
Therefore our assumption is false and h is a constant function. D
Now we prove our main result.
If Ji, f2, fa are all constant functions, the result is trivial.
Without loss of generality, suppose that Ji is nonconstant.
126 4. Solutions to Advanced Problems

For a 3 = 0, we apply the lemma to Ji and h, so h = cfi +d; for a2 = 0,

we apply the lemma to Ji and h, so h = c' Ji + d'.
Here c, c', d, d' are constant.
We have

(c'd- cd')fi + d' h - dh = (c'd- cd')fi + d'(cfi + d) - d(c' Ji+ d') = 0.

If (c' d - cd', d', -d) =I- (0, 0, 0), then let

and we are done.

Otherwise, d = d' = 0 and h, h are constant multiples of Ji.
Then the problem is again trivial.
Solution 99, Alternative 2
Define the vector
v(x) = (!1(x), h(x), h(x))
for x ER
If the v ( x) span a proper subspace of JR 3 , we cap. find a vector (c 1 , c2 , c3 )
orthogonal to that subspace, and then cif1 (x) + c2 h(x) + c3 f3(x) = 0
for all x ER
So suppose the v(x) span all of JR 3 .
Then ther~ exist X1 < x 2 < x 3 E JR such that v(x 1 ), v(x 2 ), v(x 3) are
linearly independent, and so the 3 x 3 matrix A with Aij = fj(xi) has
linearly independent rows.
But then A is invertible, and its columns also span JR 3 .
This means we can find c 1 , c2 , c3 such that
3
L ci(f,(x1), fi(x2), fi(x3)) = (0, 1, 0),
i=l

and the function cif1 +c 2 h +c3 f3 is then not monotonic, a contradiction.

4. Solutions to Advanced Problems 127

Problem 100 [USAMO 1999 proposal, Richard Stong]

Let X1, X2, ... , Xn be variables, and let Y1, Y2, ... , Y2n -1 be the sums of
nonempty subsets of Xi·
Let Pk(x1, ... , xn) be the kth elementary symmetric polynomial in
the Yi (the sum of every product-of k distinct yi's).
For which k and n is every coefficient of Pk (as a polynomial in x 1 , ... , Xn)
even?
For example, if n = 2, then Y1, Y2, y3 are x1, x2, x1 + x2 and

P1 = Y1 + Y2 + Y3 = 2x1 + 2x2,
P2 = Y1Y2 + Y2Y3 + Y3Y1 =xi+ x~ + 3x1x2,
p3 = Y1Y2Y3 = xix2 +xix~.

Solution 100
We say a polynomial Pk is even if every coefficient of Pk is even.
Otherwise, we say Pk is not even.
For any fixed positive integer n, we say a nonnegative integer k is bad
for n if k = 2n - 2j for some nonnegative integer j.
We will show by induction on n that Pk(xi, x2, · · ·, xn) is not even if and
only if k is bad for n.
For n = 1, p1(xi) = x1 is not even and k = 1 is bad for n = 1 as
k = 1 = 21 - 2° = 2n - 2°.
Suppose that the claim is true for a certain n.
We now consider Pk(x1, x2, ... , Xn+i)·
Let ak(Yi, Y2, ... , Ys) be the kth elementary symmetric polynomial.
We have the following useful, but easy to prove, facts:
1. ak(Yi, Y2, · · ·, Ys, 0) = ak(Yi, Y2, · · ·, Ys);
2. For all 1 ~ r ~ s,

O'k(Y1, ... ,ys) = L [ui(Y1,···,Yr)O'j(Yr+li···,Ys)];

i+j=k

L (x+Yi1)(x+yi 2 )···(x+yik)
ii <i2<-··<ik
k

ii<i2<· <ik r=O

00
L::: s1<s2<·· <sr
{s1,-·-,sr}~{i1,
· ,ik}
128 4. Solutions to Advanced Problems

Hence

Pk(X1,X2, · · · ,Xn+1)
= L [pi(x1,. · "Xn) ·
i+j=k
O'j(Xn+l, X1 + Xn+l, · · ·, X1 + X2 + · · · + Xn+1)]

L
i+j=k
t
( 2 ~ ~;)Pi(x1, · · ·, Xn)Pr(x1, · · ·, Xn)x~f,; ·
J
r=O

By the induction hypothesis, every term of Pr(x1, x2 · · ·, xn) is even un-

less r = 2n - 2t, for some 0 ~ t ~ n.
For such r, note that

( 2n -
j-r
r) ( 2t
- j-r
)

is even unless j - r = 0 or j - r = 2t.

Therefore, taking coefficients modulo 2,

Pk(X1,X2, · · · ,Xn+l)
= L Pi(X1, X2, · · ·, Xn)Pj(X1, X2, · · · Xn)
i+j=k

By the induction hypothesis, the terms in the first sum are even unless
k - 2n = 2n - 2u for some 0 ~ u ~ n, that is k = 2n+l - 2u.
In the second sum, every term appears twice except the term

fork even.
By the induction hypothesis, this term is even unless k /2 = 2n - 2", for
some 0 ~ v ~ n, that is k = 2n+l - 2"+ 1 .
It follows that Pk(x1, x2, · · · Xn+1) is even unless k = 2n+l - 2w for some
0 ~ w ~ n + 1, i.e., k is bad for n + 1:
4. Solutions to Advanced Problems 129

Furthermore, note that the odd coefficients in

occur for different powers of Xn+l·

Therefore, the condition that k is bad for n + 1 is also sufficient for

to be odd.
Our induction is complete.

Problem 101 [Russia 2000]

Prove that there exist 10 distinct real numbers a 1 , a2, .. ., a10 such that
the equation

(x - ai)(x - a2) · · · (x - a10) = (x + ai)(x + a2) · · · (x + a10)

has exactly 5 different real roots.
Solution 101
We show that {a 1,a2, ... ,a10} = {7,6, ... ,-2} is a group of numbers
satisfying the conditions given in the problem.
The given equality becomes

(x - 2)(x - l)x(x + l)(x + 2)g(x 2 ) = 0,

where

g(u) 2[((7 + 6 + .. · + 3)u2 +

(7 · 6 · 5 + 7 · 6 · 4 + · · · + 5 · 4 · 3)u + 7 · 6 · · · 3].

If g(u) =has no real solutions, then g(x 2) = 0 has no real solutions.

If u1 and u2 are real solutions of g( u) = 0, then u1 +u2 < 0 and u1 u2 > 0,
that is, both u 1 and u2 are negative.
It follows again that g(x 2) = 0 has no real solutions.
Our proof is complete.
 GLOSSARY

Arithmetic-Geometric Mean Inequality (AM-GM Inequality)

If a1, a2, ... , an are n nonnegative numbers, then

with equality if and only if a1 = a2 = · · · =an.

Binomial Coefficient
The coefficient of xk in the expansion of (x + l)n is

(n)k - n!
k!(n-k)!.

Cauchy-Schwarz Inequality
For any real numbers a1, a2, ... , an, and b1, b2, ... , bn

with equality if and only if ai and bi are proportional, i = 1, 2, ... , n.

Chebyshev Inequality

1. Let X1, X2 ... , Xn and Y1, Y2, ... , Yn be two sequences of real num-
bers, such that X1 ~ X2 ~ · · · ~ Xn and Y1 ~ Y2 ~ ... ~ Yn·
Then

2. Let x1, X2 ... , Xn and Y1, Y2, ... , Yn be two sequences of real num-
bers, such that X1 ~ X2 ~ · · · ~ Xn and Y1 ~ Y2 ~ ... ~ Yn·
Then
132 Glossary

De Moivre's Formula
For any angle a and for any integer n,

(cos a+ i sin a)n =cos na + i sin na.

Elementary Symmetric Polynomials (Functions)

Given indeterminates x 1 , ... , Xn, the elementary symmetric functions
s 1 , ... , Sn are defined by the relation (in another indeterminate t)

(t + X1) · · (t + Xn) = tn + S1tn-l +···+Sn-It+ Sn.

That is, Sk is the sum of the products of the x, taken k at a time. It

is a basic result that every symmetric polynomial in x 1 , ... , Xn can be
(uniquely) expressed as a polynomial in the Si, and vice versa.

Fibonacci Numbers
Sequence defined recursively by F1 = F2 = 1, Fn+2 = Fn+I + Fn, for all
nEN.

Jensen's Inequality
If f is concave up on an interval [a, b] and .>.1, .>.2, ... , An are nonnegative
numbers with sum equal to 1, then

for any x 1 , x2, ... , Xn in the interval [a, b]. If the function is concave
down, the inequality is reversed.

Lagrange's Interpolation Formula

Let xo, x1, ... , Xn be distinct real numbers, and let yo, Y1, ... , Yn bear-
bitrary real numbers. Then there exists a unique polynomial P(x) of
degree at most n such that P(xi) = Yi, i = 0, 1, ... , n. This is the
polynomial given by

P( x ) -_ Ln y,. (x - xo) · · · (x - Xi-1)(x - Xi+I) · · · (x - Xn) .

i=O (xi - Xo) ... (xi - Xi-I)(xi - Xi+I) ... (xi - Xn)

Law of Cosines
Let ABC be a triangle. Then

BC 2 = AB 2 + AC2 - 2AB · AC cos A.

Glossary 133

Lucas' Theorem
Let p be a prime; let a and b be two positive integers such that

where 0 ~ ai, bi < pare integers for i = 0, 1, ... , k. Then

Pigeonhole Principle
If n objects are distributed among k < n boxes, some box contains at
least two objects.

Root Mean Square-Arithmetic Mean Inequality (RMS-AM In-

equality)
For positive numbers xi, x2, ... , Xn,

J x~ + x~ +n · · · + x~
-=-~-=-~~~--= > xi + x2 + ·· · + Xk .
- n

More generally, let ai,a2, ... ,an be any positive numbers for which ai +
a2 +···+an= 1. For positive numbers xi, x2, ... , Xn we define

M_ 00 = min{x1, X2, ... , xk},

Moo= max{x1, x2, ... , Xk},
Mo = x~i x~2 ... x~n,
Mt = (a1xi + a2x~ + · · · + akxt) 11t,
where tis a non-zero real number. Then

for s ~ t.
134 Glossary

Triangle Inequality
Let z =a+ bi be a complex number. Define the absolute value of z to
be
!z! = Ja2 + b2.
Let a and f3 be two complex numbers. The inequality
!a + f31 ~ !al + lf31

is called the triangle inequality.

Let a = a1 + a2i and {3 = f31 + f32i, where a1, a2, f31, f32 are real numbers.
Then a+ f3 = (a1 + f31) + (a2 + f32)i.
Vectors u = [a1, a2], v = [f31, f32], and w = [a1 + f31, a2 + f32] form a
triangle with sides lengths !a!, !f3!, and !a+ f31.
The triangle inequality restates the fact that the length of any side of a
triangle is less than the sum of the lengths of the other two sides.

Vieta's Theorem
Let x 1, x2, ... , Xn be the roots of polynomial

P(x) = anXn + an-1Xn-l + ... + a1x + ao.

where an "I 0 and ao, a1, ... , an E CC. Let Sk be the sum of the products
of the xi taken k at a time. Then
k an-k
Sk = (- l) - - ,
an
that is,
an-1
X1 + X2 + · · · + Xn = - - - ;
an
an-2
X1X2 + · · · + XiXj + Xn-lXn = - - ;
an

Trigonometric Identities

sin2 a+ cos 2 a = 1,
sin a
tanx=--,
cos a
1
cotx = - - ' - -
tan a
Glossary 135

addition and subtraction formulas:

sin( a± b) =sin a cos b ±cos a sin b,

cos( a± b) =cos a cos b =F sin a sin b,
tana( ± b) = tana ± tanb ;
l=i=tanatanb

double-angle formulas:
sin2a = 2sinacosa,
cos 2a = cos 2 a - sin 2 a = 2 cos 2 a - 1 = 1 - 2 sin 2 a,
2tana
tan2a= 1 ,
-tan2 a

triple-angle formulas:
sin3a = 3sina - 4sin3 a,
cos 3a = 4 cos3 a - 3 cos a,
3 _ 3 tan a - tan a.
3
tan a- 1 - 3 tan 2 a '

half-angle formulas:
. 2tan~
sma= 2 11 ,
l+tan 2
1 - tan 2 $!
cosa= 1
+tan 2
2 ~,
_ 2tan~ .
tana- 1 211 ,
-tan 2

sum-to-product formulas:
. . b . a+b a-b
sma + sm = 2 sm- 2 -cos- 2-,
a+b a-b
cosa + cosb = 2cos- 2 -cos - 2 -;
sin( a+ b)
tana + tanb = ;
cosacosb

difference-to-product formulas:
. . b . a-b a+b
sma-sm = 2 sm- 2-cos-
2 -,
136 Glossary

. a-b . a+b
cos a - cos b = - 2 sm - - sm - -
2 2 '
sin( a - b)
tana-tanb= b;
cos a cos

product-to-sum formulas:

2 sin a cos b =sin( a+ b) +sin( a - b),

2cosacosb = cos(a + b) + cos(a - b),
2sinasinb = - cos(a + b) + cos(a - b).
 FURTHER READING

1. Andreescu, T. Kedlaya, K.; Zeitz, P., Mathematical Contests

1995-1996: Olympiad Problems from around the World, with
Solutions, American Mathematics Competitions, 1997.

2. Andreescu, T. Kedlaya, K., Mathematical Contests 1996-1997:

Olympiad Problems from around the World, with Solutions,
American Mathematics Competitions, 1998.

3. Andreescu, T. Kedlaya, K., Mathematical Contests 1997-1998:

Olympiad Problems from around the World, with Solutions,
American Mathematics Competitions, 1999.

4. Andreescu, T. Feng, Z., Mathematical Olympiads: Problems and

Solutions from around the World, 1998-1999, Mathematical
Association of America, 2000.

5. Andreescu, T. Gelca, R., Mathematical Olympiad Challenges,

Birkhiiuser, 2000.

6. Barbeau, E., Polynomials, Springer-Verlag, 1989.

7. Beckenbach, E. F., Bellman, R., An Introduction to Inequalities,

New Mathematical Library, Vol. 3, Mathematical Association of
America, 1961.

8. Chinn, W. G., Steenrod, N. E., First Concepts of Topology, New

Mathematical Library, Vol. 27, Random House, 1966.

9. Cofman, J., What to Solve?, Oxford Science Publications, 1990.

10. Coxeter, H. S. M., Greitzer, S. L., Geometry Revisited, New

Mathematical Library, Vol. 19, Mathematical Association of
America, 1967.

11. Doob, M., The Canadian Mathematical Olympiad 1969-1993,

University of Toronto Press, 1993.

12. Engel, A., Problem-Solving Strategies, Problem Books in

Mathematics, Springer, 1998.

13. Fomin, D., Kirichenko, A., Leningrad Mathematical Olympiads

1987-1991, MathPro Press, 1994.
138 Further Reading

14. Fomin, D., Genkin, S., Itenberg, I., Mathematical Circles,

American Mathematical Society, 1996.

15. Graham, R. L., Knuth, D. E., Patashnik, 0., Concrete

Mathematics, Addison-Wesley, 1989.

16. Greitzer, S. L., International Mathematical Olympiads, 1959-1977,

New Mathematical Library, Vol. 27, Mathematical Association of
America, 1978.

17. Grossman, I., Magnus, W., Groups and Their Graphs, New
Mathematical Library, Vol. 14, Mathematical Association of
America, 1964.

18. Kazarinoff, N. D., Geometric Inequalities, New Mathematical

Library, Vol. 4, Random House, 1961.

19. Klamkin, M., International Mathematical Olympiads, 1978-1985,

New Mathematical Library, Vol. 31, Mathematical Association of
America, 1986.

20. Klamkin, M., USA Mathematical Olympiads, 1972-1986, New

Mathematical Library, Vol. 33, Mathematical Association of
America, 1988.

21. Kiirschak, J., Hungarian Problem Book, Volumes I & II, New
Mathematical Library, Vols. 11 & 12, Mathematical Association of
America, 1967.

22. Kuczma, M., 144 Problems of the Austrian-Polish Mathematics

Competition 1978-1993, The Academic Distribution Center, 1994.

23. Larson, L. C., Problem-Solving Through Problems, Springer-Verlag,

1983.
24. Lausch, H., Bosch Giral, C., The Asian Pacific Mathematics Olym-
piad 1989-2000, AMT Publishing, Canberra, 2001.

25. Liu, A., Chinese Mathematics Competitions and Olympiads

1981-1993, AMT Publishing, Canberra, 1998.

26. Lozansky, E., Rousseau, C. Winning Solutions, Springer, 1996.

27. Ore, 0., Graphs and their uses, Random House, 1963.

28. Ore, 0., Invitation to number theory, Random House, 1967.

29. Sharygin, I. F., Problems in Plane Geometry, Mir, Moscow, 1988.

Further Reading 139

30. Sharygin, I. F., Problems in Solid Geometry, Mir, Moscow, 1986.

31. Shklarsky, D. 0, Chentzov, N. N; Yaglom, I. M., The USSR
Olympiad Problem Book, Freeman, 1962.
32. Slinko, A., USSR Mathematical Olympiads 1989-1992, AMT
Publishing, Canberra, 1997.
33. Soifer, A., Colorado Mathematical Olympiad: The first ten years,
Center for excellence in mathematics education, 1994.
34. Szekely, G. J., Contests in Higher Mathematics, Springer- Verlag,
1996.
35. Stanley, R. P., Enumerative Combinatorics, Cambridge University
Press, 1997.
36. Taylor, P. J., Tournament of Towns 1980-1984, AMT Publishing,
Canberra, 1993.
37. Taylor, P. J., Tournament of Towns 1984-1989, AMT Publishing,
Canberra, 1992.
38. Taylor, P. J., Tournament of Towns 1989-1993, AMT Publishing,
Canberra, 1994.
39. Taylor, P. J., Storozhev, A., Tournament of Towns 1993-1997,
AMT Publishing, Canberra, 1998.
40. Tomescu, I., Problems in Combinatorics and Graph Theory, Wiley,
1985.
41. Vanden Eynden, C., Elementary Number Theory, McGraw-Hill,
1987.
42. Wilf, H. S., Generatingfunctionology, Academic Press, 1994.
43. Wilson, R., Introduction to graph theory, Academic Press, 1972.
44. Yaglom, I. M., Geometric Transformations, New Mathematical
Library, Vol. 8, Random House, 1962.
45. Yaglom, I. M., Geometric Transformations II, New Mathematical
Library, Vol. 21, Random House, 1968.
46. Yaglom, I. M., Geometric Transformations III, New Mathematical
Library, Vol. 24, Random House, 1973.
47. Zeitz, P., The Art and Craft of Problem Solving, John Wiley &
Sons, 1999.
 Titu Andreescu is
Director of the
Amer i can
Mathema t ics
Compe titio ns,
serves as Head
Coach of the USA
l nternat iona l
Ma th em a ti ca l
Olympiad (IMO)
Team, is Chair of the USA Ma thematical
Olympiad Committee, and is Director of the
USA Mathematical Olympiad Summer
Program. Originally from Romania, Prof.
Andreescu received the Distinguished
Professor Award from the Romanian
Ministry of Education in 1983, then after
moving lo the tJSA was awarclrcl thr F.clyth
May Sliffe Award for Distinguished High
School Mathematics Teaching fro m the
MAA in 1994. ln addition, he received a
Certificate of Appreciation presented by the
President of the MAA fo r "his outstanding
service as coach of the USA Mathematical
Olympiad Program in preparing the USA
team for its perfect performance in Hong
Kong at the 1994 IMO".

Zuming Feng
graduated with a
PhD from Johns
Hopkins University
with an emphasis
on Algebraic
Number Theory
and Elli ptic
Curves. He teaches
at Phillips Exeter
Academy. 1le also serves as a coach of the
USA International Mathematical Olympiad
{IMO) Team, is a member of the USA
Mathematical Olympiad Committee, and is
an assistan t director of the USA
Mathematical Olympiad Summer Program.
He received the Edyth May Sliffe Award for
Disti nguished High School Mathematics
TH E A USTRALI AN MHH EMATI CS TR UST Teaching fro m the MAA in 1996.

E NR I C H ME~T S E R IES ISBN I 876420 12 X