2003-04.wh-ley

Improved Perfect Slice Sampling

Wolfgang Hörmann and Josef Leydold


Abstract

Perfect slice sampling is a method to turn Markov Chain Monte Carlo (MCMC) samplers into exact generators for independent random variates. The originally proposed method is rather slow and thus several improvements have been suggested. However, two of them are erroneous. In this article we give a short introduction to perfect slice sampling, point out incorrect methods, and give a new improved version of the original algorithm.


Mathematics Subject Classification: 65C05 (Monte Carlo Methods) 65C10 (Random Number Generation)

CR Categories and Subject Descriptors: G.3 [Probability and Statistics]: Random number generation

General Terms: Algorithms

Key Words: random variate generation, Markov chain Monte Carlo method, perfect sampling, coupling from the past, slice sampler, layered multishift coupling


Download Preprint


Josef.Leydold@statistik.wu-wien.ac.at