Advanced Markov Chain Monte Carlo Methods: Learning from Past Samples

Developing algorithms that are immune to the local trap problem has long been considered as the most important topic in MCMC research. Various advanced MCMC algorithms which address this problem have been developed include, the modified Gibbs sampler, the methods based on auxiliary variables and the methods making use of past samples. The focus of this book is on the algorithms that make use of past samples.

This book includes the multicanonical algorithm, dynamic weighting, dynamically weighted importance sampling, the Wang-Landau algorithm, equal energy sampler, stochastic approximation Monte Carlo, adaptive MCMC algorithms, conjugate gradient Monte Carlo, adaptive direction sampling, the sampling Metropolis-Hasting algorithm and the multiplica sampler.