prev | colloquium schedule

Colloquium - 4 May 2000

Department of Statistics
Texas A&M University

ABSTRACT: In survival analysis, the relationship between a survival time and covariates is conveniently modeled with the proportional hazards regression model. This model usually assumes that covariates have log-linear effects on the hazard function. We consider the proportional hazards regression model with an unspecified nonparametric link function and unspecified baseline hazard function. A two-stage iterative algorithm is proposed to estimate the covariate effects and the unknown link function, as well as the baseline hazard function. The procedure is illustrated through simulations and data sets.

For longitudinal studies that involve time-dependent covariates, both the aforementioned two-stage procedure and its asymptotic properties can be adapted to accommodate time-dependent covariates, provided that the entire history of the covariates is observed. In reality, this is often not the case, as the sampling times for each subject may vary. We will discuss how to cope with such irregular sampling designs, and propose a new procedure motivated by functional data analysis methodology.

This is joint work with Wei Wang (PhD student).

Refreshments will be served in the Blocker Building, Room 447, at 3:30 p.m.

Copyright © Unless otherwise stated, all contents of these web pages are copyright by their respective authors or by the Department of Statistics, Texas A&M University, 1995-2001. Comments and questions about this site should be sent to web@stat.tamu.edu.