We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.CO

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Mathematics > Combinatorics

Title: A Stirling-type formula for the distribution of the length of longest increasing subsequences, applied to finite size corrections to the random matrix limit

Abstract: The discrete distribution of the length of longest increasing subsequences in random permutations of order $n$ is deeply related to random matrix theory. In a seminal work, Baik, Deift and Johansson provided an asymptotics in terms of the distribution of the largest level of the large matrix limit of GUE. As a numerical approximation, however, this asymptotics is inaccurate for small lengths and has a slow convergence rate, conjectured to be just of order $n^{-1/3}$. Here, we suggest a different type of approximation, based on Hayman's generalization of Stirling's formula. Such a formula gives already a couple of correct digits of the length distribution for $n$ as small as $20$ but allows numerical evaluations, with a uniform error of apparent order $n^{-3/4}$, for $n$ as large as $10^{12}$; thus closing the gap between a table of exact values (that has recently been compiled for up to $n=1000$) and the random matrix limit. Being much more efficient and accurate than Monte-Carlo simulations for larger $n$, the Stirling-type formula allows for a precise numerical understanding of the first few finite size correction terms to the random matrix limit, a study that has recently been initiated by Forrester and Mays, who visualized the form of the first such term. We display also the second one, of order $n^{-2/3}$, and derive (heuristically) expansions of expected value and variance of the length, exhibiting several more terms than previously known.
Comments: 25 pages, 6 figures, 3 tables. V4: Fig. 4 with $n=10^{10}$; improved discussion of accuracies in {\S}4; enhanced least squares fits for mean and variance. V3: variance discussed in new {\S}4.4. V2: simplifications in {\S}3.2: derived kernels have finite rank; improved numerical stabilization in {\S}3.3; associated table of exact values has been extended for up to $n=1000$
Subjects: Combinatorics (math.CO); Mathematical Physics (math-ph); Numerical Analysis (math.NA); Probability (math.PR)
MSC classes: 05A16, 60B20, 30D15, 47N40
Cite as: arXiv:2206.09411 [math.CO]
  (or arXiv:2206.09411v4 [math.CO] for this version)

Submission history

From: Folkmar Bornemann [view email]
[v1] Sun, 19 Jun 2022 14:18:50 GMT (332kb,D)
[v2] Tue, 28 Jun 2022 09:02:56 GMT (307kb,D)
[v3] Thu, 30 Jun 2022 16:17:23 GMT (308kb,D)
[v4] Mon, 4 Jul 2022 09:09:45 GMT (305kb,D)
[v5] Mon, 26 Sep 2022 13:43:15 GMT (307kb,D)
[v6] Mon, 24 Oct 2022 08:03:29 GMT (307kb,D)
[v7] Mon, 14 Nov 2022 15:45:15 GMT (309kb,D)

Link back to: arXiv, form interface, contact.