English

Markov chains with heavy-tailed increments and asymptotically zero drift

Probability 2019-06-24 v1

Abstract

We study the recurrence/transience phase transition for Markov chains on R+\mathbb{R}_+, R\mathbb{R}, and R2\mathbb{R}^2 whose increments have heavy tails with exponent in (1,2)(1,2) and asymptotically zero mean. This is the infinite-variance analogue of the classical Lamperti problem. On R+\mathbb{R}_+, for example, we show that if the tail of the positive increments is about cyαc y^{-\alpha} for an exponent α(1,2)\alpha \in (1,2) and if the drift at xx is about bxγb x^{-\gamma}, then the critical regime has γ=α1\gamma = \alpha -1 and recurrence/transience is determined by the sign of b+cπcosec(πα)b + c\pi \textrm{cosec} (\pi \alpha). On R\mathbb{R} we classify whether transience is directional or oscillatory, and extend an example of Rogozin \& Foss to a class of transient martingales which oscillate between ±\pm \infty. In addition to our recurrence/transience results, we also give sharp results on the existence/non-existence of moments of passage times.

Keywords

Cite

@article{arxiv.1806.07166,
  title  = {Markov chains with heavy-tailed increments and asymptotically zero drift},
  author = {Nicholas Georgiou and Mikhail V. Menshikov and Dimitri Petritis and Andrew R. Wade},
  journal= {arXiv preprint arXiv:1806.07166},
  year   = {2019}
}
R2 v1 2026-06-23T02:34:30.646Z