BAYES AND EMPIRICAL BAYES METHODS FOR DATA ANALYSIS, 2nd ED.

               by Bradley P. Carlin and Thomas A. Louis
         Boca Raton, FL:  Chapman and Hall/CRC Press, June 2000

             Texts in Statistical Science (the "red series")
      419 pp., 50 illustrations, hardback   ISBN 1-58488-170-4
                         2000 price: US $54.95

BRIEF DESCRIPTION: 

In recent years, Bayes and empirical Bayes (EB) methods have continued to
increase in popularity and impact.  Building on the first edition of their
popular text, in this second edition Carlin and Louis introduce these
methods, demonstrate their usefulness in challenging applied settings,
and show how they can be implemented using modern Markov chain Monte
Carlo (MCMC) methods.   Their presentation is accessible to those new
to Bayes and empirical Bayes methods, while providing in-depth coverage.

In addition, this new edition features an updated guide to Bayesian
software (such as WinBUGS) with worked examples, and provides
comprehensive subject and author indices.  Other enhancements new to
this edition include:

  ... a less technical introductory chapter comparing Bayes and
        frequentist inference, with motivating examples
  ... a gentler introduction to Gibbs sampling and full conditional
        distributions 
  ... several recent developments in MCMC, such as reversible jump
        MCMC, slice sampling, structured MCMC, and overrelaxation
  ... explicit description of how to estimate MCMC standard errors
  ... an expanded and revised treatment of Bayesian model choice
  ... new material on several methodological areas, such as
         spatial statistics, sequential analysis, and sample size
         estimation for clinical trials
  ... a new decision theory appendix
  ... more illustrations, exercises, and solutions
  ... a completely updated reference section

With its broad appeal as a text for those in biomedical science,
education, social science, agriculture, and engineering, this second
edition offers a relatively gentle and comprehensive introduction for
statisticians and practitioners already familiar with more traditional
frequentist statistical methods.  Focusing on practical tools for data
analysis, the book shows how properly structured Bayes and EB
procedures typically have good frequentist and Bayesian performance,
both in theory and in practice.

READERSHIP:  The book will be of interest to -- 

* practicing statisticians and biostatisticians seeking to learn new
methods for analyzing complex datasets from industry, agriculture,
biomedical science, public health, and other fields;
 
* statistical consultants looking to broaden their toolkit of
procedures for challenging datasets;

* instructors wanting to incorporate recent developments in Bayesian
methods into their courses on nonlinear models, longitudinal data
analysis, time series, survival analysis, spatial statistics, and so
on;

* statistical researchers in application areas needing an introduction
to the benefit of Bayesian methods for hierarchical modeling;

* Masters and Ph.D. students in statistics and biostatistics
whose departments offer a course on Bayes and/or EB methods (usually
in the 2nd year, after the basic math stat course has been completed);

* Masters and Ph.D. students in statistics, biostatistics,
epidemiology, and related fields who recognize the need for advanced
modeling tools in their thesis research, but whose departments do not
offer formal coursework on Bayes and EB methods.

----------------------------------------------------------------------

Bradley P. Carlin is Professor in the Division of Biostatistics in the
School of Public Health at the University of Minnesota, Minneapolis,
MN, USA.  Thomas A. Louis is Senior Statistician, The RAND Corporation,
Santa Monica, CA, USA.