JSM 2004 - Toronto

Abstract #302105

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: 383
Type: Contributed
Date/Time: Wednesday, August 11, 2004 : 2:00 PM to 3:50 PM
Sponsor: ENAR
Abstract - #302105
Title: Combining Dependent Tests to Compare the Accuracies of Different Diagnostic Tests
Author(s): Yuqing Yang*+
Companies: Columbia University
Address: 154 Haven Ave. #707, New York, NY, 10032,
Keywords: ROC curve ; repeated measurements ; combining dependent tests
Abstract:

In comparing different diagnostic tests, repeated measurements of the same characteristic are often taken under two or more distinct conditions for each experimental subject. The commonly seen example is the multireader ROC studies with multiple tests. To draw an overall conclusion on whether different diagnostic tests have same accuracies, various parametric and nonparametric methods have been proposed. We present a method of combining dependent tests by utilizing the idea in Wei & Johnson. Specifically, a test statistic from each reader is constructed by extending the structural components method of DeLong, DeLong, and Clark-Pearson, then a single test is constructed by combining all the individual test statistics. Numerical studies show that when readers have effect on the measurements, the proposed test yields appropriate size and power while tests based on existing method have inappropriate size and insufficient power.

  • 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