JSM 2011 Online Program

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: 653
Type: Topic Contributed
Date/Time: Thursday, August 4, 2011 : 10:30 AM to 12:20 PM
Sponsor: Section on Bayesian Statistical Science
Abstract - #301976
Title: Bayesian Analysis of Two-Phase Studies with Binary Outcome and Categorical Covariates
Author(s): Michelle Ross*+ and Jon Wakefield
Companies: University of Washington and University of Washington
Address: 200 17th Ave E #104, Seattle, WA, 98112,
Keywords: Bayesian statistics ; Two-phase studies ; Wilms Tumor ; Markov Chain Monte Carlo
Abstract:

Two-phase sampling is an important outcome-dependent stratification technique that was developed to increase the efficiency of parameter estimation. To date, many frequentist solutions have been proposed, however no Bayesian solution has been offered. These frequentist solutions can encounter difficulties when there are small cell counts present (for example, in the context of a rare disease and a rare exposure) since these methods rely on asymptotic results for inference. It is also not possible to simply incorporate random effects in these approaches. Hence, there is strong motivation for a Bayesian approach to the analysis of two-phase studies. In this paper, we develop methodology for the Bayesian analysis of two-phase studies with a binary outcome and categorical covariates. We use a Markov Chain Monte Carlo auxiliary variable algorithm to implement the approach. The practicality of these methods is illustrated and compared to previously established methods using data from the National Wilms Tumor Study.


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.