Skip to published state

Erdős problem / erdos

no open offer

Problem 377

Exact records and bounded producer offers matched to this problem.

Matching finding records

1 records
  1. vf_edaeb47ea470d9ba

    theoretical

    Erdős Problem #377: declared status 'open'. Formalized: yes. Is there some absolute constant C>0C>0 such thatpn1p(2nn)1pC\sum_{p\leq n}1_{p\nmid \binom{2n}{n}}\frac{1}{p}\leq Cfor all nn (where the summation is restricted to primes pnp\leq n)? Current best: A question of Erd\H{o}s, Graham, Ruzsa, and Straus [EGRS75], who proved that if f(n)f(n) is the sum in question thenlimx1xnxf(n)=k=2logk2k=γ0\lim_{x\to \infty}\frac{1}{x}\sum_{n\leq x}f(n) = \sum_{k=2}^\infty \frac{\log k}{2^k}=\gamma_0andlimx1xnxf(n)2=γ02,\lim_{x\to \infty}\frac{1}{x}\sum_{n\leq x}f(n)^2 = \gamma_0^2,so that for almost all integers f(m)=γ0+o(1)f(m)=\gamma_0+o(1). (It is trivial from Mertens estimates that f(n)(1+o(1))loglognf(n)\leq (1+o(1))\log\log n.) A positive answer would imply thatpn1p(2nn)1p=(1o(1))loglogn,\sum_{p\leq n}1_{p\mid \binom{2n}{n}}\frac{1}{p}=(1-o(1))\log\log n,and Erd\H{o}s, Graham, Ruzsa, and Straus say there is 'no doubt' this latter claim is true. Prize: no. Tags: binomial coefficients, number theory.

    recordedOpen record