|
Activity Number:
|
111
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #301334 |
|
Title:
|
Bayesian Analysis of Covariate Profiles
|
|
Author(s):
|
John T. Molitor*+ and Michail Papathomas and Sylvia Richardson
|
|
Companies:
|
Imperial College, London and Imperial College, London and Imperial College, London
|
|
Address:
|
Division of Epidemiology, Public Health and Primary Care, London, W2 1PG, United Kingdom
|
|
Keywords:
|
profiles ; children's health ; correlated data ; bayesian analysis ; mcmc
|
|
Abstract:
|
Standard regression analyses are often plagued with problems that occur when one tries to make meaningful inference using datasets that contain a large number of correlated variables. In this manuscript, we propose an inferential data analysis method that uses, as its basic unit of inference, a profile, formed from a sequence of covariate values. The model presented is based on Bayesian partition models. Our implementation of this approach extends the standard partition model in a number of important ways, such as, a) allowing number of clusters to be random, b) performing variable selection, and c) utilizing a set of post-processing procedures to provide an examination and comparison of different partitions of the data. An analysis of children's health data from The National Survey of Children's Health (NSCH) is provided.
|
- 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 2008 program |
