Online Program Home
My Program

Abstract Details

Activity Number: 430
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #320479 View Presentation
Title: Multivariate Copula-Based Regression Models for Longitudinal Data
Author(s): Xin Tian* and Colin O. Wu
Companies: National Heart, Lung, and Blood Institute and National Heart, Lung, and Blood Institute
Keywords: longitudinal analysis ; multivariate distribution ; copula ; nonparametric modeling ; time-varying transformation models ; dependence

In biomedical studies, it is an important challenge to estimate joint distribution of multivariate longitudinal outcome variables, where marginal outcome variables may have non-normal distributions that change with time and time-dependent covariates. We propose to use multivariate copulas to link marginal outcomes and model their dependence, while individual marginal distributions conditioning on longitudinal covariates may be estimated by a class of flexible time-varying transformation models. An application of these models is to jointly track multivariate risk factors over time. Our models and estimation method are demonstrated through an epidemiological study of childhood growth and health study.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association