The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Abstract Details
|
Activity Number:
|
618
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 4, 2011 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #302964 |
|
Title:
|
Bayesian Testing in ANOVA Using Neutral-Data Comparisons
|
|
Author(s):
|
Dan J. Spitzner*+
|
|
Companies:
|
University of Virginia
|
|
Address:
|
P.O. Box 400135, Charlottesville, VA, 22904-4135,
|
|
Keywords:
|
ANOVA ;
Bayes factors ;
default priors ;
Bayesian consistency ;
partition analysis ;
Bayesian hypothesis testing
|
|
Abstract:
|
It is common for Bayesian approaches to classical analysis problems such as ANOVA to focus on estimation rather than testing. Part of the reason is that the default priors for estimation may be understood in a straightforward way, whereas the underlying intuition of default priors for testing is complex and subtle. In this talk I will discuss a new testing procedure which gives sensible results even when using vague priors that would be specified for estimation. This is a ``neutral-data comparison,'' which serves as both an alternative to the Bayes factor, and an approximation to it. The result for Bayesian ANOVA is a coherent, flexible multiple-testing tool that supplements the usual estimation-based analysis. Among its benefits, the approach provides sophisticated, modern tools with which to test multiple contrasts, allowing, e.g., the classical ``mean separation'' procedure to evolve into a comprehensive partition analysis.
|
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 2011 program
|
2011 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.