9. A Primer in Bayesian Data Analysis

  1. James R. Thompson

Published Online: 29 NOV 2011

DOI: 10.1002/9781118109656.ch9

Empirical Model Building: Data, Models, and Reality, Second Edition

Empirical Model Building: Data, Models, and Reality, Second Edition

How to Cite

Thompson, J. R. (2011) A Primer in Bayesian Data Analysis, in Empirical Model Building: Data, Models, and Reality, Second Edition, John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9781118109656.ch9

Publication History

  1. Published Online: 29 NOV 2011
  2. Published Print: 24 OCT 2011

ISBN Information

Print ISBN: 9780470467039

Online ISBN: 9781118109656

SEARCH

Keywords:

  • Bayesian data analysis;
  • data augmentation algorithm;
  • EM algorithm;
  • Gibbs sampler

Summary

There are many situations where Bayesian analysis is very useful. This chapter gives some examples. The EM algorithm is an iterative procedure whereby the hypothesized failure time can be conjectured and the log likelihood reformulated for another attempt to obtain a new estimate for the modal value. The chapter shows the application of the EM in the analysis of the times of remission of leukemia patients using a new drug and those using an older modality of treatment. The data used are from a clinical trial designed by Gehan and Freireich. The database has been used by Cox and Oakes as an example of the EM algorithm. The chapter uses it to examine the EM algorithm, data augmentation, chained data augmentation, and the Gibbs sampler. Unlike the EM algorithm, there is no maximization step in data augmentation, rather a series of expectations.

Controlled Vocabulary Terms

EM algorithm