Unofficial archive — problems, solutions & results © IMC, reproduced with permission.

IMC / 2002 / Problems / Day 1, P2

IMC 2002 · Day 1 · P2

medium

Does there exist a continuously differentiable function f:RRf : \mathbb{R} \to \mathbb{R} such that for every xRx \in \mathbb{R} we have f(x)>0f(x) > 0 and f(x)=f(f(x))f'(x) = f(f(x))?

Solution (official)

Assume that there exists such a function. Since f(x)=f(f(x))>0f'(x) = f(f(x)) > 0, the function is strictly monotone increasing.

By the monotonity, f(x)>0f(x) > 0 implies f(f(x))>f(0)f(f(x)) > f(0) for all xx. Thus, f(0)f(0) is a lower bound for f(x)f'(x), and for all x<0x < 0 we have f(x)<f(0)+xf(0)=(1+x)f(0)f(x) < f(0) + x \cdot f(0) = (1 + x) f(0). Hence, if x1x \le -1 then f(x)0f(x) \le 0, contradicting the property f(x)>0f(x) > 0.

So such function does not exist.

How the field did

contestants scored
182
average (of 20)
12.31
solved (≥ 80%)
48.9%
near-0 (≤ 10%)
16.5%
discrimination
0.53

Score distribution (field cohort)

Computed on contestants with a meaningful total (field cohort); discrimination is the corrected item–total correlation.

Similar problems

IMC 2008 · Day 1 · P1mediumavg 5.4/10 · solved 48% · near-0 40% · disc 0.42
IMC 2023 · Day 1 · P1easyavg 7.7/10 · solved 72% · near-0 6% · disc 0.46
IMC 2006 · Day 2 · P8easyavg 8.2/10 · solved 76% · near-0 4% · disc 0.41
IMC 2018 · Day 1 · P4hardavg 1.7/10 · solved 16% · near-0 81% · disc 0.48