A unified framework is used here for representing and understanding Gaussian Markov random fields (GMRFs). Case studies illustrate the use of GMRFs in complex hierarchical models in which statistical inference is only possible using Markov chain Monte Carlo techniques. Emphasis is on the connection between GMRFs and numerical methods for sparse matrices. A C-library for fast and exact simulation is provided. Author information is not given. Annotation ©2004 Book News, Inc., Portland, OR (booknews.com)