Use of the Gibbs sampler in expert systems

✍ Scribed by Jeremy York

• Publisher: Elsevier Science
• Year: 1992
• Tongue: English
• Weight: 87 KB
• Volume: 56
• Category: Article
• ISSN: 0004-3702
• DOI: 10.1016/0004-3702(92)90032-s

✦ Synopsis

In my paper "Use of the Gibbs sampler in expert systems", I attempted to survey statistics and genetics literature of interest to researchers in AII faded to stress, however, that the best reference for understanding Markov chain Monte Carlo (MCMC) methods is the paper by Hastings [3 ], which generalizes the Metropolis algorithm [7] Several recent papers, and others which I neglected to reference, are described below In early May 1992, papers on MCMC were read by Smith and Roberts [ 10 ], Besag and Green [1 ], and Gilks et al [2] at a meeting of the Royal Statistical Society. The first of these has the most general discussion, and illustrates that the Gibbs sampler IS a special case of the methods presented in [3] Anyone wishing to understand the theoretical issues involved is well advised to consult the thorough and rigorous paper by Tlerney [11] or the abbreviated version [ 12 ] In the genetics literature, Ott [ 8 ] and Ploughman and Boehnke [ 9 ] present methods for drawing independent samples from the distribution of interest if the graphical structure is simple Lange and Matthysse [4] and Lange