Abstract #301346


The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.


Back to main JSM 2002 Program page



JSM 2002 Abstract #301346
Activity Number: 272
Type: Topic Contributed
Date/Time: Wednesday, August 14, 2002 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Computing*
Abstract - #301346
Title: Fitting Algorithms for Empirical Bayes Analysis of Microarray Data
Author(s): Brian Caffo*+
Affiliation(s): Johns Hopkins University
Address: 615 N Wolfe Street, Baltimore, Maryland, 21205, USA
Keywords: microarray ; EM algorithm ; Monte Carlo
Abstract:

We explore fitting algorithms for obtaining hyperparameters for empirical Bayesian modeling of microarray data. This setting is particularly complicated because of the computational burden arising from the often extremely large number of genes under study. We focus our attention on the EM algorithm and its variants. When estimating hyperparameters with the EM algorithm, considerable simplification occurs, allowing for separate maximizing of clusters of hyperparameters (for the models under consideration). Further, the "maximization'' step of the EM algorithm relies only on a few sums and, hence, only one pass through the gene index is required per EM iteration. Extensions to situations with non-conjugate prior distributions where the "expectation" step must be approximated via Monte Carlo or Markov chain Monte Carlo will also be discussed.


  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2002 program

JSM 2002

For information, contact meetings@amstat.org or phone (703) 684-1221.

If you have questions about the Continuing Education program, please contact the Education Department.

Revised March 2002