Analysis of a nonreversible Markov chain sampler

• 成果类型: Article
• 署名作者: Diaconis, P; Holmes, S; Neal, RM
• 署名单位: Stanford University; INRAE; University of Toronto; University of Toronto
• 刊物名称: ANNALS OF APPLIED PROBABILITY
• ISSN/ISSBN: 1050-5164
• 发表日期: 2000
• 页码: 726-752
• 关键词: monte-carlo; Metropolis algorithm; GIBBS SAMPLER; distributions; relaxation; walk

摘要

We analyze the convergence to stationarity of a simple nonreversible Markov chain that serves as a model for several nonreversible Markov chain sampling methods that are used in practice. Our theoretical and numerical results show that nonreversibility can indeed lead to improvements over the diffusive behavior of simple Markov chain sampling schemes. The analysis uses both probabilistic techniques and an explicit diagonalization.