Likelihood Methods in Statistics / Edition 1

by Thomas A. Severini

Overview

This book provides an introduction to the modern theory of likelihood-based statistical inference. This theory is characterized by several important features. One is the recognition that it is desirable to condition on relevant ancillary statistics. Another is that probability approximations are based on saddlepoint and closely related approximations that generally have very high accuracy. A third aspect is that, for models with nuisance parameters, inference is often based on marginal or conditional likelihoods, or approximations to these likelihoods. These methods have been shown often to yield substantial improvements over classical methods. The book also provide an up-to-date account of recent results in the field, which has been undergoing rapid development.

Product Details

ISBN-13:	9780198506508
Publisher:	Oxford University Press
Publication date:	01/18/2001
Series:	Oxford Statistical Science Series , #22
Edition description:	New Edition
Pages:	392
Product dimensions:	9.32(w) x 6.32(h) x 1.08(d)

About the Author

Northwestern University

Table of Contents

1. Some basic concepts2. Large-sample approximations3. Likelihood4. First-order asymptotic theory5. Higher-order asymptotic theory6. Asymptotic theory and conditional inference7. The signed likelihood ratio statistic8. Likelihood functions for a parameter of interest9. The modified profile likelihood functionAppendix: Data sets used in the examplesReferencesAuthor indexSubject index

Editorial Reviews

"This book presents an excellent overview of modern likelihood methods. The presentation is clear and readable enough to be of considerable use to intermediate level graduate students, but is complete and up-to-date enough to act as an excellent first reference for beginning researchers in the field. . . . Moreover, there are excellent discussions of important references as well as numerous exercises. . . . The field of likelihood asymptotics has been undergoing an increasingly rapid development over the last two decades, and this book provides an excellent and detailed account of where the field is, where it came from and perhaps even a bit about where it is going." — Mathematical Reviews

From the Publisher

This book is suitable for students and statisticians with a knowledge of graduate-level statistical theory. At least some background in classical likelihood theory is recommended, but the opening chapters provide a review for readers who lack such background. Severini (statistics, Northwestern University) intends his book as an introduction to the modern, alternative approach to likelihood inference. The text includes a discussion of asymptotic theory and conditional inference; the signed likelihood ratio statistic; likelihood functions for a parameter of interest; and the modified profile likelihood function. Each chapter includes extensive references to additional results, followed by exercises to illustrate concepts from the chapter. Solutions are not included in the text but are available from the author (e-mail address included in the Preface). Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews