Publisher's Synopsis
This is a unified presentation of a variety of computational algorithms used in likelihood and Bayesian inference. In this second edition, Martin Tanner has taken the opportunity to expand the treatment of many of the techniques discussed, to devote more space to comparing the methods covered, and to describe the applications in more detail. Topics covered include: maximum likelihood, Monte Carlo methods, the EM algorithm, data augmentation techniques, imputation methods, the Gibbs sampler, the Metropolis algorithm, and the griddy Gibbs sampler.;The reader is assumed to have a reasonable basic background in statistics as might be gained in the first year of a graduate course, but otherwise the book is self-contained. As a result, the book will provide a survey of the fast-moving area of statistics for research statisticians and for other researchers and graduate students whose research touches on these techniques.
