English

Random walks with echoed steps I

Probability 2025-10-31 v2

Abstract

A random walk with echoed steps (RWES) is a process {S~n}n1={X~1++X~n}n1\{\tilde{S}_n\}_{n\geq1}=\{\tilde{X}_1+\cdots+\tilde{X}_n\}_{n\geq1} that inserts memory and echo into an ordinary random walk (ORW) with i.i.d. steps, X1++XnX_1+\cdots+X_n. The RWES is defined recursively as follows. Let S~1=X1\tilde{S}_1=X_1. With probability 1p1-p, the nn-th increment of the RWES follows that of the ORW, X~n=Xn\tilde{X}_n=X_n. Otherwise, X~n\tilde{X}_n is set as a random echo of a uniform sample of the past steps X~1,,X~n1\tilde{X}_1,\dots,\tilde{X}_{n-1} determined by a random factor ξn\xi_n. Namely, X~n=ξnX~U[n]\tilde{X}_n=\xi_n\tilde{X}_{\mathcal{U}[n]} with probability pp, where U[n]\mathcal{U}[n]\simUniform{1,,n1}\{1,\dots,n-1\}. The RWES is a broad generalization of the elephant random walk and of the positively/negatively/unbalanced step-reinforced random walks. We determine strong convergences of S~\tilde{S} when the echo law ξ\xi is non-negative. The rates of convergence are determined by the product pEξp\mathbb{E}\xi and exhibit a phase transition with critical value at pEξ=1p\mathbb{E}\xi=1. Highlight that in its super-critical regime, the RWES has super-linear scaling exponents --observed for the first time in this type of random walks with memory--. We provide Laws of Large Numbers, conditions for the convergence of S~\tilde{S} around its mean towards random series and provide some distributional properties of the limits. Our approach relies on the interpretation of the model in terms of continuous time branching random walks, random recursive trees, P\'olya urns, and associated martingales.

Keywords

Cite

@article{arxiv.2510.24881,
  title  = {Random walks with echoed steps I},
  author = {Daniela Portillo del Valle},
  journal= {arXiv preprint arXiv:2510.24881},
  year   = {2025}
}

Comments

45 pages, 1 figure, corrected typos

R2 v1 2026-07-01T07:10:27.941Z