Regular Article
What Do We Know about the Metropolis Algorithm ?,☆☆

https://doi.org/10.1006/jcss.1998.1576Get rights and content
Under a Creative Commons license
open archive

Abstract

The Metropolis algorithm is a widely used procedure for sampling from a specified distribution on a large finite set. We survey what is rigorously known about running times. This includes work from statistical physics, computer science, probability, and statistics. Some new results (Propositions 6.1–6.5) are given as an illustration of the geometric theory of Markov chains.

Cited by (0)

Both authors research partially supported by NSF Grant DMS 9504379 and Nato Grant CRG 950686

☆☆

G. Grimmett