JSM 2004 - Toronto

Abstract #301473

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2004 Program page
Activity Number: 200
Type: Contributed
Date/Time: Tuesday, August 10, 2004 : 9:00 AM to 10:50 AM
Sponsor: Section on Bayesian Statistical Science
Abstract - #301473
Title: Bayesian Meta-analysis of ROC with Normal Mixtures
Author(s): Minje Sung*+ and Alaattin Erkanli and Refik Soyer
Companies: Duke University Medical Center and Duke University Medical Center and George Washington University
Address: DUMC 3454, Durham, NC, 27710,
Keywords: Dirichlet mixture model ; receiver operating characteristics (ROC) ; Bayesian hierarchical model ; Markov chain Monte Carlo simulation
Abstract:

We develop a semiparametric Bayesian approach to multivariate receiver operating characteristics (ROC) using mixtures of Dirichlet Process (DP) priors. We address two challenging issues: modeling nonstandard distribution of screen scores and combining results from several different such ROC studies. In medical testing the results of a screening test usually does not have a Gaussian or a symmetric distribution. If this is the case, using a normal distribution will lead to incorrect choices of the diagnostic cut-offs and unreliable estimates of prevalence of a disease. The DP mixtures would provide a robust way of modeling nonstandard distributions. Hierarchical modeling framework is proposed to obtain study-specific ROCs as well as combined ROC. The model is implemented with real data from child psychiatric studies using Markov chain Monte Carlo technique.

  • 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 2004 program

JSM 2004 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.
Revised March 2004