Subset Selection in Regression / Edition 2

by Alan Miller

Overview

Originally published in 1990, the first edition of Subset Selection in Regression filled a significant gap in the literature, and its critical and popular success has continued for more than a decade. Thoroughly revised to reflect progress in theory, methods, and computing power, the second edition promises to continue that tradition. The author has thoroughly updated each chapter, incorporated new material on recent developments, and included more examples and references.

New in the Second Edition:

A separate chapter on Bayesian methods

Complete revision of the chapter on estimation

A major example from the field of near infrared spectroscopy

More emphasis on cross-validation

Greater focus on bootstrapping

Stochastic algorithms for finding good subsets from large numbers of predictors when an exhaustive search is not feasible

Software available on the Internet for implementing many of the algorithms presented

More examples

Subset Selection in Regression, Second Edition remains dedicated to the techniques for fitting and choosing models that are linear in their parameters and to understanding and correcting the bias introduced by selecting a model that fits only slightly better than others. The presentation is clear, concise, and belongs on the shelf of anyone researching, using, or teaching subset selecting techniques.

Product Details

ISBN-13:	9781584881711
Publisher:	Taylor & Francis
Publication date:	04/15/2002
Series:	Chapman & Hall/CRC Monographs on Statistics and Applied Probability , #95
Edition description:	REV
Pages:	256
Product dimensions:	6.00(w) x 9.00(h) x (d)

About the Author

Miller, Alan

Table of Contents

Introduction. Least-Squares Computations. Finding Subsets which Fit Well. Hypothesis Testing. How Many Variables? Estimation after Model Selection. Bayesian Methods. Conclusions.

Editorial Reviews

Miller (Commonwealth Scientific & Industrial Research Organization, Australia), focusing almost entirely on nonlinear regression, presents a monograph that collects what is known about estimation techniques and discusses some new material. It is the aim of the author to provide information on the proper empirical choice of a model, given a set of data and many variables or alternative models from which to select. This new edition adds material on Bayesiean methods, as well as stochastic algorithms for finding good subsets from large numbers of predictors. Annotation c. Book News, Inc., Portland, OR (booknews.com)

Monograph intended to alert professional statisticians and advanced students of the field to a class of problems which arise from the too- routine application of the methods of linear regression, and to the existence of techniques which permit some of those problems to be circumvented. Seven chapters, with references and examples, but no exercises. (NW) Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews