Estimation of Rank-Tracking Probabilities Using Nonparametric Mixed-Effects Models for Longitudinal Data

• Activity Number: 156
• Type: Invited
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Korean International Statistical Society
• Author(s): Colin O. Wu* and Xin Tian
• Companies: NIH and National Heart, Lung, and Blood Institute
• Keywords: Basis approximation ; Conditional distribution ; Longitudinal study ; Mixed model ; Time-varying coefficient model ; Rank-tracking probability

Abstract

An important scientific objective of longitudinal studies involves tracking the probability of a subject having certain health status over the course of the study. Proper definitions and estimates of disease risk tracking have important implications in the design and analysis of long-term biomedical studies and in developing guidelines for disease prevention and intervention. We study in this paper a class of "rank-tracking probabilities" (RTP) to describe a subject's conditional probabilities of having certain health outcomes at two different time points. Structural nonparametric estimation and inferences for the RTPs and their functions are developed based on nonparametric mixed-effects models and B-spline smoothing methods. Statistical properties of our procedures are investigated through a simulation study. We apply our methods to an epidemiological study of childhood cardiovascular risk factors, and demonstrate that the RTPs and their nonparametric estimators provide useful tools to quantitatively evaluate whether the cardiovascular risks, such as obesity and hypertension, can be tracked from early childhood to adolescence.