← All problems

Erdős Problem #692 — Unimodality of Divisor Densities

1 year old · traced to

For positive integers n<mn<m, let

E(n,m)={x∈N:#{d∈N:n<d<m, d∣x}=1},E(n,m)=\left\{x\in\mathbb N:\#\{d\in\mathbb N:n<d<m,\ d\mid x\}=1\right\},

and let δ1(n,m)\delta_1(n,m) be its natural density, namely

δ1(n,m)=lim⁡X→∞1X#{x∈E(n,m):1≤x≤X}.\delta_1(n,m)=\lim_{X\to\infty}\frac{1}{X}\#\{x\in E(n,m):1\le x\le X\}.

(The set is periodic, so this density exists.) For fixed nn, is the sequence (δ1(n,m))m≥n+2\bigl(\delta_1(n,m)\bigr)_{m\ge n+2} unimodal: does there exist an index M≥n+2M\ge n+2 such that it is nondecreasing for n+2≤m≤Mn+2\le m\le M and nonincreasing for m≥Mm\ge M? The question is whether this holds for every positive integer nn. Erdős Problem #692 Resolution of Erdős’ problems about unimodularity (arXiv:2501.10333v1)

Background

Erdős asked this question at Oberwolfach in 1986. He had proved a uniform upper bound δ1(n,m)≪(log⁡n)−c\delta_1(n,m)\ll (\log n)^{-c} for all mm, for some constant c>0c>0; Ford later obtained sharper bounds in various ranges of nn and mm. Those estimates concern how small the density is, whereas unimodality asks how it changes as the upper endpoint moves. Erdős Problem #692

References

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.