This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 343
Type: Contributed
Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
Sponsor: Biometrics Section
Abstract - #308610
Title: Concordance Measures: Estimation Methods and Potential Bias Due to Censoring
Author(s): Norberto Pantoja*+ and Rebecca Betensky
Companies: Harvard School of Public Health/FDA and Harvard School of Public Health
Address: , , ,
Keywords: C-index ; ROC curve ; Alzheimer's disease
Abstract:

A concordance probability is a measure of predictive accuracy of a marker for an event, which occurs at random times and may be right censored. We review different estimation procedures for this measure presented in the literature. An important issue is how censored times are handled. Some of these procedures make model assumptions and perform well under such assumptions. Other methods are nonparametric but do not use all the available information, therefore bias is present, particularly under high levels of censoring. Uno et al (2009) proposes a nonparametric method that includes inverse probability weights in their estimator. Such estimator assumes that the censoring is independent of the marker. We extend their result and propose an estimator that performs well under the assumption of censoring dependent of the marker. Applications on Alzheimer's disease are presented.


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 2010 program




2010 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.