For a polynomial P∈R[x], let f(P)=n if n is the smallest positive integer such that (∀x∈R)n(P(P(…P(x))…))>0, and f(P)=0 if such an integer n does not exist. Does there exist a polynomial P∈R[x] of degree 20142015 such that f(P)=2015?
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.
The answer is that it does exist such a polynomial. Actually we shall prove a more general result: Let s be an even integer and t>1 be an arbitrary integer. Then for some constant c>0 for the polynomial P(x)=(x+1)s+c−1 (which is of degree s) we have f(P)=t. Indeed: The polynomial P is strictly increasing function on the interval [−1,∞). Let xk(c) be the minimal value of the polynomial kP(P(…P(x))…) for a fixed c>0,
Consider xk(c) is strictly increasing, because of x1(c)=c−1>−1 and xk+1(c)=P(xk(c)). The equation xt−1(c)=0 (where c is the unknown). Since the leading coefficient of the polynomial xt−1(c) equals 1 and the constant term is −1 (we can prove these claims trivially by induction on t), this polynomial has a positive zero. Let c0 be one of them. Now for the polynomial P(x)=(x+1)s+c0−1 we have xt−1(c0)=0, and therefore xt(c0)=P(0)=c0>0. This completes the proof. □
Source: MathNet,
licensed CC-BY-4.0.
Statement reproduced verbatim; metadata (topic, difficulty, ordering) added by this project.