Poisson limits for large boxes in sparse random allocations

Let $(Y_1,\dots,Y_n)$ be the random allocation of $m$ balls into $n$ boxes in which an allocation $(y_1,\dots,y_n)$ has probability proportional to $\prod_iw_{y_i}$, and let $\xi_1,\dots,\xi_n$ be independent with $\mathbb P(\xi_i=k)$ proportional to $w_k\tau^k$, where $\tau$ is chosen so that $\mathbb E\xi_i=m/n$. Janson showed that when $m/n$ converges to a positive limit below a threshold determined by the weights, the numbers of boxes with many balls are asymptotically Poisson and the largest boxes of the allocation are close in distribution to the largest of $\xi_1,\dots,\xi_n$, and he asked whether the same holds when $m/n\to0$. Let $a$ be the least $k\ge1$ with $w_k>0$. The answer is positive for every sequence with $m\to\infty$ if every $k$ with $w_k>0$ is a multiple of $a$, for instance if $w_1>0$, and it is negative along every sequence on which $m$ is bounded. If some $k$ with $w_k>0$ is not a multiple of $a$, there are sequences with $m\to\infty$ and $m/n\to0$ along which the answer is negative. When the least such $k$, say $b$, is coprime to $a$, we characterize the sequences along which the answer is positive in terms of $n\mathbb P(\xi_1=b)$ and the residue of $m$ modulo $a$; when $n\mathbb P(\xi_1=b)$ has a positive finite limit, the number of boxes with at least $b$ balls converges to a Poisson distribution conditioned on a residue class. The proofs rest on an exact formula for the joint law of the boxes whose sizes are neither $0$ nor $a$ and on a local limit theorem for the binomial distribution with explicit constants.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-08
DOI
https://doi.org/10.5281/zenodo.24069924
Primary Topic
Probability and Statistical Research
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Poisson limits for large boxes in sparse random allocations

Yitzchak Shmalo
Zenodo (CERN European Organization for Nuclear Research)
Probability and Statistical Research
preprint

Poisson limits for large boxes in sparse random allocations

Yitzchak Shmalo
preprint en

Abstract

Let $(Y_1,\dots,Y_n)$ be the random allocation of $m$ balls into $n$ boxes in which an allocation $(y_1,\dots,y_n)$ has probability proportional to $\prod_iw_{y_i}$, and let $\xi_1,\dots,\xi_n$ be independent with $\mathbb P(\xi_i=k)$ proportional to $w_k\tau^k$, where $\tau$ is chosen so that $\mathbb E\xi_i=m/n$. Janson showed that when $m/n$ converges to a positive limit below a threshold determined by the weights, the numbers of boxes with many balls are asymptotically Poisson and the largest boxes of the allocation are close in distribution to the largest of $\xi_1,\dots,\xi_n$, and he asked whether the same holds when $m/n\to0$. Let $a$ be the least $k\ge1$ with $w_k>0$. The answer is positive for every sequence with $m\to\infty$ if every $k$ with $w_k>0$ is a multiple of $a$, for instance if $w_1>0$, and it is negative along every sequence on which $m$ is bounded. If some $k$ with $w_k>0$ is not a multiple of $a$, there are sequences with $m\to\infty$ and $m/n\to0$ along which the answer is negative. When the least such $k$, say $b$, is coprime to $a$, we characterize the sequences along which the answer is positive in terms of $n\mathbb P(\xi_1=b)$ and the residue of $m$ modulo $a$; when $n\mathbb P(\xi_1=b)$ has a positive finite limit, the number of boxes with at least $b$ balls converges to a Poisson distribution conditioned on a residue class. The proofs rest on an exact formula for the joint law of the boxes whose sizes are neither $0$ nor $a$ and on a local limit theorem for the binomial distribution with explicit constants.

Zenodo (CERN European Organization for Nuclear Research)
Hebrew University of Jerusalem (IL)
Probability and Statistical Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon