Bayesian Statistics

An Introduction

Third Edition

PETER M. LEE

ISBN 0-340-81405-5

Preface to the Third Edition

I have been gratified by the reception of previous editions of this book and have tried to respond to comments from reviewers and readers in preparing the third edition.

Two chapters were added to the second edition. Chapter 8 deals with hierarchical methods in Bayesian statistics and gives a fuller treatment of empirical Bayes methods. Some of the material in this chapter originated in lectures given at a summer school organized by the Società Italiana di Statistica in Varese in September 1995. Chapter 9 deals with real numerical methods, especially the EM algorithm and Gibbs sampling. In the present edition, this chapter has been expanded by a fuller treatment of the Metropolis-Hastings algorithm in Section 9.6, an introduction to WinBUGS in Section 9.7 and a brief treatment of generalized linear models in Section 9.8.

In addition, a treatment of Bernardo’s theory of reference distributions has been added as Section 3.11. The R project continues to increase in popularity, and accordingly the programs formerly given in “pseudo-code” in Chapter 9 are now given in R. Further, Appendix C which gives R functions for finding highest density regions for the commoner distributions (as well as functions concerned with Behrens’ distribution) has been added. The website associated with the book,

http://www.york.ac.uk/depts/maths/histstat/pml1/bayes/book.htm

(note that in the above pml are letters followed by the digit 1), works through all numerical examples in R as well as giving solutions to all the exercises in the book.

Peter M. Lee

27 October 2003
Click on the required option below: