English

Ergodicity of Markov chain Monte Carlo with reversible proposal

Methodology 2016-02-10 v1 Statistics Theory Statistics Theory

Abstract

We describe ergodic properties of some Metropolis-Hastings (MH) algorithms for heavy-tailed target distributions. The analysis usually falls into sub-geometric ergodicity framework but we prove that the mixed preconditioned Crank-Nicolson (MpCN) algorithm has geometric ergodicity even for heavy-tailed target distributions. This useful property comes from the fact that the MpCN algorithm becomes a random-walk Metropolis algorithm under suitable transformation.

Keywords

Cite

@article{arxiv.1602.02889,
  title  = {Ergodicity of Markov chain Monte Carlo with reversible proposal},
  author = {Kengo Kamatani},
  journal= {arXiv preprint arXiv:1602.02889},
  year   = {2016}
}

Comments

14 pages