Semiparametric Estimation Methods for Longitudinal Data with Informative Observation Times

• Activity Number: 160
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Risk Analysis
• Author(s): Xingqiu Zhao*+
• Companies: The Hong Kong Polytechnic University
• Keywords: Asymptotic normality ; Empirical process ; Estimating equation ; Informative observation process ; Polynomial spline

Abstract

Analyzing irregularly spaced longitudinal data often involve the response process and the observation process where the two processes may be correlated even given covariates. In this talk we propose a new class of semiparametric mean models which allows for the interaction between the observation history and covariates, leaving patterns of the observation process to be arbitrary. For inference on the regression parameters and the baseline mean function, a spline-based least square estimation approach is proposed, and the consistency, rate of convergence and asymptotic normality of the proposed estimators are established. Simulation studies demonstrate that the proposed inference procedure performs well. The analysis of a bladder tumor data and a medical cost data are presented to illustrate the proposed method.

Research supported in part by grants from HKSAR-RGC-GRF.