|
Activity Number:
|
453
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #301240 |
|
Title:
|
Bayesian Nonparametric Polya Tree Mixture Models with Application to Random Effects Meta-Analysis
|
|
Author(s):
|
Adam Branscum*+ and Timothy Hanson
|
|
Companies:
|
University of Kentucky and The University of Minnesota
|
|
Address:
|
College of Public Health, Lexington, KY, 40536,
|
|
Keywords:
|
Bayesian nonparametrics ; Polya trees
|
|
Abstract:
|
A goal in meta-analysis is estimation of a single effect measure using data from several studies that are each designed to address the same scientific inquiry. Because studies are typically conducted in geographically disperse locations, the statistical analysis of meta-analytic data often involves the use of random effects models that account for study-to-study variability. To eliminate the influence of overly restrictive parametric models on inferences, we develop a novel hierarchical Bayesian nonparametric Polya tree mixture model. We present methodology for testing the Polya tree mixture versus a normal model. An application involving meta-analysis of epidemiologic studies designed to characterize the association between alcohol consumption and breast cancer is presented, which together with results from simulated data highlight the performance of Polya tree mixtures.
|
