Factoring Factorial n

From MaRDI portal





The authors discuss the following old question of P. Erdős: Let us express \(n!\) as a product of \(n\) positive integers, the smallest of which is as large as possible, i.e. if \(n!= a_1a_2\cdots a_n\), \(a_1\leq a_2\leq\cdots\leq a_n\), determine the largest \(a_1\) possible. By Stirling's formula \(a_1\leq (n/e)(1+ o(1))\), and over 15 years ago Erdős, Selfridge and Strauss announced that they had a proof that this was the best possible, but the proof was lost. The authors could prove that, for sufficiently large \(n\), \(n!\) can be expressed as the product of \(n\) factors, each greater than \(n/4\) and they conjectured that apart from the case \(n= 56\), \(\max a_1\geq [2n/7]\), and for \(n\geq 300,000\), \(\max a_1\geq n/3\).











This page was built for publication: Factoring Factorial n

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4222427)