Maths Olympiad Prep

Track / Stage 6 / 242 of 400 #1242 of 1964

Problem 1242

National Olympiad, first round
Algebra Difficulty 6.3 Prove it The 36th KOREAN MATHEMATICAL OLYMPIAD Final Round · South Korea

Let R+\mathbb{R}^+ be the set of positive real numbers. Let f:R+R+f : \mathbb{R}^+ \to \mathbb{R}^+ be a function satisfying the following.
For each positive real number xx, there exists yRy \in \mathbb{R} such that
(x+f(y))(y+f(x))4, (x + f(y))(y + f(x)) \leq 4,
and the number of such yy's is finite.
Prove that f(x)>f(y)f(x) > f(y) for every pair of positive real numbers xx and yy with x<yx < y.

This one wants a proof. Work it on paper, then read the official solution and mark yourself. Be honest about it: the record is only any use to you if it is.

Next problem →

Official solution

For each positive real number xx, let AxA_x be the set of positive real numbers yy satisfying (x+f(y))(y+f(x))4(x + f(y))(y + f(x)) \le 4. The following are easy consequences by the definition.
(1) If xAyx \in A_y, then yAxy \in A_x.
(2) For x<yx < y, if f(x)f(y)f(x) \le f(y) then AyAxA_y \subseteq A_x.
We prove the following lemma.
Lemma. For xR+x \in \mathbb{R}^+, there are only finitely many positive real numbers y(<x)y (< x) such that f(y)<f(x)f(y) < f(x).
Proof. Since AxA_x is not empty, there exists zAxz \in A_x. For y(<x)y (< x) with f(y)<f(x)f(y) < f(x), we have AxAyA_x \subseteq A_y by (2). So zz also belongs to AyA_y, and yAzy \in A_z by (1). Therefore, if there are infinitely many y(<x)y (< x) such that f(y)<f(x)f(y) < f(x), AzA_z becomes an infinite set, which is a contradiction. \square
Assume that there exist a,bR+a, b \in \mathbb{R}^+ such that a<ba < b and f(a)f(b)f(a) \le f(b). Let S={xa<x<b,f(x)>f(a)}S = \{x \mid a < x < b, f(x) > f(a)\}. By the lemma, there are only finitely many xx with a<x<ba < x < b such that f(x)<f(b)f(x) < f(b). So SS is an infinite set. For any sSs \in S, AsAaA_s \subseteq A_a by (2). Since AsA_s is not empty, there exists tAsAat \in A_s \subseteq A_a. So, sAtzAaAzs \in A_t \subseteq \bigcup_{z \in A_a} A_z by (1). That is, SzAaAzS \subseteq \bigcup_{z \in A_a} A_z which yields a contradiction because SS is an infinite set but
zAaAzzAaAz | \bigcup_{z \in A_a} A_z | \le \sum_{z \in A_a} |A_z|
is finite. \square

Source: MathNet, licensed CC-BY-4.0. Statement reproduced verbatim; metadata (topic, difficulty, ordering) added by this project.