Student Seminar Series - September 13, 2004
University of Minnesota
School of Statistics
College of Liberal Arts

Comparison of Kaplan-Meier-type and Self-Consistent Estimators of the Bivariate Survivor Function

Luu Pham


Monday, September 13, 2004
10:00 AM, TBA Ford Hall
Minneapolis, East Bank Campus

Refreshments at 9:30 AM
300 Ford Hall

Abstract

 Many non-parametric estimators of the bivariate survival function for right-censored data have been proposed. One approach is an explicit estimator derived as a product-limit or Kaplan Meier-type estimator, which produces a non-negligible proportion of negative mass points. A second approach is an implicit estimator which is based on the solution to Efron's Self-Consistency equation, and depends on bandwidths. I review the univariate estimators, and provide descriptions of the computation of the bivariate estimators. Simulation study results follow. A distance measure is proposed for comparing the estimators.