Large Deviations and Convexity Transition for the Multidimensional Elephant Random Walk

In this article we establish a weak large deviation principle for the empirical measures of the elephant random walk on $\mathbb{Z}^d$, for every reinforcement parameter $p \in (0,1)$. We identify a qualitative change in the behaviour of the associated rate function between dimension $d=1$ and dimensions $d \geq 2$. In dimension $d=1$, the rate function coincides with the Donsker--Varadhan rate function of the simple symmetric random walk and is independent of the reinforcement parameter $p$. By contrast, in every dimension $d\geq 2$, the rate function differs from that of the simple random walk as soon as $p\neq 1/(2d)$. Finally, we prove that for all $d \geq 2$ the rate function is convex if and only if $p \leq 1/(2d)$. This allows us to identify a critical point in the large-deviation behaviour of the model which occurs strictly before the standard diffusive-to-superdiffusive crossover.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-18
DOI
https://doi.org/10.48550/arxiv.2611.24132
Primary Topic
Stochastic processes and statistical mechanics
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Large Deviations and Convexity Transition for the Multidimensional Elephant Random Walk

Jan-Luka Fatras
arXiv (Cornell University)
Stochastic processes and statistical mechanics
preprint

Large Deviations and Convexity Transition for the Multidimensional Elephant Random Walk

Jan-Luka Fatras
preprint en

Abstract

In this article we establish a weak large deviation principle for the empirical measures of the elephant random walk on $\mathbb{Z}^d$, for every reinforcement parameter $p \in (0,1)$. We identify a qualitative change in the behaviour of the associated rate function between dimension $d=1$ and dimensions $d \geq 2$. In dimension $d=1$, the rate function coincides with the Donsker--Varadhan rate function of the simple symmetric random walk and is independent of the reinforcement parameter $p$. By contrast, in every dimension $d\geq 2$, the rate function differs from that of the simple random walk as soon as $p\neq 1/(2d)$. Finally, we prove that for all $d \geq 2$ the rate function is convex if and only if $p \leq 1/(2d)$. This allows us to identify a critical point in the large-deviation behaviour of the model which occurs strictly before the standard diffusive-to-superdiffusive crossover.

arXiv (Cornell University)
Stochastic processes and statistical mechanics
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