Functional Association Models for Multivariate Survival Processes

• Activity Number: 393
• Type: Invited
• Date/Time: Thursday, August 12, 2004 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Jun Yan*+ and Jason Fine
• Companies: University of Iowa and University of Wisconsin, Madison
• Keywords: empirical process ; functional estimating equation ; local dependence ; partially observed ; uniform convergence ; varying-coefficient

Abstract

We consider multivariate temporal processes that are continuously observed within overlapping time windows. The intended application is censored multistate and multivariate survival settings, where point processes are continuously observed. Functional mean and association regression models are studied for the point processes, with completely unspecified time-varying coefficients. The continuous observation scheme is exploited: the coefficients may be estimated nonparametrically by extending GEE to continuously observed data. The estimators automatically converge at the parametric rate, without smoothing, unlike with discretely observed data. Uniform consistency and weak convergence is established with empirical process techniques. Existing functional approaches to survival processes utilize intensity models, which require smoothing and depend critically on the choice of smoothing parameters, similarly to discretely observed data. The nonparametric estimators yield new tests for covariate effects, parametric submodeling of these effects, and goodness-of-fit testing. Simulation studies and an analysis of familial aggregation of alcoholism illustrate the methodology's utility.