English

A Zero-One Law for Markov Chains

Probability 2020-11-10 v1

Abstract

We prove an analog of the classical Zero-One Law for both homogeneous and nonhomogeneous Markov chains (MC). Its almost precise formulation is simple: given any event AA from the tail σ\sigma-algebra of MC (Zn)(Z_n), for large nn, with probability near one, the trajectories of the MC are in states ii, where P(AZn=i)P(A|Z_n=i) is either near 00 or near 11. A similar statement holds for the entrance σ\sigma-algebra, when nn tends to -\infty. To formulate this second result, we give detailed results on the existence of nonhomogeneous Markov chains indexed by Z\mathbb Z_- or Z\mathbb Z in both the finite and countable cases. This extends a well-known result due to Kolmogorov. Further, in our discussion, we note an interesting dichotomy between two commonly used definitions of MCs.

Keywords

Cite

@article{arxiv.2011.04063,
  title  = {A Zero-One Law for Markov Chains},
  author = {Michael Grabchak and Isaac Sonin},
  journal= {arXiv preprint arXiv:2011.04063},
  year   = {2020}
}
R2 v1 2026-06-23T19:59:43.770Z