Partially Linear Hazard Regression for Multivariate Survival Data

• Activity Number: 229
• Type: Invited
• Date/Time: Tuesday, August 8, 2006 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Jianwen Cai*+ and Jianqing Fan and Jiancheng Jiang and Haibo Zhou
• Companies: The University of North Carolina at Chapel Hill and Princeton University and Princeton University and The University of North Carolina at Chapel Hill
• Keywords: local pseudo-partial likelihood ; marginal hazard model ; martingale ; multivariate failure time ; partially linear ; profile pseudo-partial likelihood

Abstract

Partially linear hazard regression models for multivariate survival data are studied. A profile pseudo-partial likelihood estimation method is proposed under the marginal hazard model framework. The estimation on the parameters for the linear part is accomplished via maximization of a profile pseudo-partial likelihood. This enables one to get root-n consistent estimators of the parametric component. Asymptotic normality is obtained for the estimates of both the linear and the nonlinear parts. The new technical challenge is that the nonparametric component is indirectly estimated via its integrated derivative function from a local polynomial fit. An algorithm of fast implementation of our proposed method is presented. Consistent standard error estimates are proposed. Simulations are conducted to examine the performance of the proposed method. A real dataset is analyzed for illustration.