Connects many different aspects of the growing model selection field by examining different lines of reasoning that have motivated derivation of both classical and modern criteria, and then examining the performance of these criteria to see how well it matches with the intent of their creators. Useful as a guide to researchers, and as a resource for practicing statisticians for matching appropriate selection criteria to a given problem or data set. Contains chapters on univariate regression and autoregressive models, cross-validation and the bootstrap, robust regression and quasi-liklihood, and nonparametric regression and wavelets. The authors are affiliated with North Dakota State University, and the University of California-Davis. Annotation c. Book News, Inc., Portland, OR (booknews.com)