English

Ergodic Theorems for discrete Markov chains

Probability 2017-08-01 v2

Abstract

Let XnX_n be a discrete time Markov chain with state space SS (countably infinite, in general) and initial probability distribution μ(0)=(P(X0=i1),P(X0=i2),,)\mu^{(0)} = (P(X_0=i_1),P(X_0=i_2),\cdots,). What is the probability of choosing in random some kNk \in \mathbb{N} with knk \leq n such that Xk=jX_k = j where jSj \in S? This probability is the average 1nk=1nμj(k)\frac{1}{n} \sum_{k=1}^n \mu^{(k)}_j where μj(k)=P(Xk=j)\mu^{(k)}_j = P(X_k = j). In this note we will study the limit of this average without assuming that the chain is irreducible, using elementary mathematical tools. Finally, we study the limit of the average 1nk=1ng(Xk)\frac{1}{n} \sum_{k=1}^n g(X_k) where gg is a given function for a Markov chain not necessarily irreducible.

Keywords

Cite

@article{arxiv.1707.08827,
  title  = {Ergodic Theorems for discrete Markov chains},
  author = {Nikolaos Halidias},
  journal= {arXiv preprint arXiv:1707.08827},
  year   = {2017}
}
R2 v1 2026-06-22T20:59:05.787Z