English

On joint returns to zero of Bessel processes

Probability 2024-06-28 v1

Abstract

In this article, we consider joint returns to zero of nn Bessel processes (n2n\geq 2): our main goal is to estimate the probability that they avoid having joint returns to zero for a long time. More precisely, considering nn independent Bessel processes (Xt(i))1in(X_t^{(i)})_{1\leq i \leq n} of dimension δ(0,1)\delta \in (0,1), we are interested in the first joint return to zero of any two of them: Hn:=inf{t>0,1i<jn such that Xt(i)=Xt(j)=0}. H_n := \inf\big\{ t>0, \exists 1\leq i <j \leq n \text{ such that } X_t^{(i)} = X_t^{(j)} =0 \big\} \,. We prove the existence of a persistence exponent θn\theta_n such that P(Hn>t)=tθn+o(1)\mathbb{P}(H_n>t) = t^{-\theta_n+o(1)} as tt\to\infty, and we provide some non-trivial bounds on θn\theta_n. In particular, when n=3n=3, we show that 2(1δ)θ32(1δ)+f(δ)2(1-\delta)\leq \theta_3 \leq 2 (1-\delta) + f(\delta) for some (explicit) function f(δ)f(\delta) with sup[0,1]f(δ)0.079\sup_{[0,1]} f(\delta) \approx 0.079.

Cite

@article{arxiv.2406.19344,
  title  = {On joint returns to zero of Bessel processes},
  author = {Quentin Berger and Loïc Béthencourt and Camille Tardif},
  journal= {arXiv preprint arXiv:2406.19344},
  year   = {2024}
}

Comments

26 pages, 3 figures, comments are welcome!

R2 v1 2026-06-28T17:21:41.332Z