General Departmental Seminar Series
Nonparametric Comparison of Survival Functions Based on Interval Censored Data
Tony Sun, Department of Statistics,
University of Missouri-Columbia
Friday, March 2, 2001 at 12:00-1:00 p.m.
3285 Medical Sciences Center - 1300 University Ave.
Interval-censored failure time data often occur in biological and medical studies such as medical follow-up studies and clinical trials. In this situation one problem of great interest to investigators is the treatment comparison based on interval-censored data and for this purpose, a few nonparametric and semiparametric tests have been proposed. In this talk, we will first review the existing methods, most of which are rank-based tests and not sensitive to nonproportional types of treatment differences, and then present some alternative tests. Simulation studies were performed to evaluate and compare the proposed methods with existing methods and suggest that the proposed tests work well for nonproportional alternatives as well as proportional alternatives. An illustrative example is discussed.
Back to General Departmental Seminar Series