The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
373
|
Type:
|
Invited
|
Date/Time:
|
Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Bayesian Statistical Science
|
Abstract - #303719 |
Title:
|
Bayesian Testing of Associations and Interactions with Multivariate Categorical Data
|
Author(s):
|
David Dunson*+ and Anirban Bhattacharya
|
Companies:
|
Duke University and Duke University
|
Address:
|
Box 90251, Durham , NC, , USA
|
Keywords:
|
Bayesian ;
Nonparametrics ;
Contingency table ;
High-dimensional ;
Genetics ;
Categorical data
|
Abstract:
|
Massive-dimensional and enormously sparse contingency table data are routinely collected in many application areas. For example, in genetic epidemiology studies it is common to collect data on single nucleotide polymorphisms (SNPs) at many different loci along with demographic and exposure information. SNPs can be categorized as 0,1,2 while other variables are also typically categorized, leading to high-dimensional categorical data that can be organized into a contingency table. Often in epidemiology and other settings there may be high-dimensional interactions (e.g., among SNPs and among SNPs and environmental factors). Such interactions are notoriously difficult to detect and it becomes impractical to use traditional methods for contingency table analysis, such as log-linear models, since even with two-way interactions the number of terms in such models can blow up beyond the ability of current computers. Graphical modeling approaches face similar problems. To address these challenges, we propose new classes of nonparametric Bayesian models for the higher-order probability tensor characterizing the joint distribution of the categorical data. Adding to the literature on ten
|
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 2012 program
|
2012 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.