Activity Number:
|
594
- Recent Advances in Statistical Modeling for Multivariate/Correlated/Time-Varying Longitudinal Data
|
Type:
|
Invited
|
Date/Time:
|
Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
|
Sponsor:
|
WNAR
|
Abstract #300418
|
Presentation
|
Title:
|
Unified Multivariate Longitudinal Analysis Using Dynamic Copula Models
|
Author(s):
|
Wei Zhang and Colin O. Wu* and Xin Tian and Qizhai Li
|
Companies:
|
Eunice Kennedy Shriver National Institute of Child Health and Human Development, NIH and National Heart, Lung and Blood Institute, National Institutes of Health and National Heart, Lung and Blood Institute, National Institutes of Health and Academy of Mathematics and Systems Science, Chinese Academy of Science
|
Keywords:
|
Basis approximation;
Conditional distribution function;
Dynamic copula model;
Multivariate longitudinal data;
Structured nonaprametric regression;
Time-varying parameter
|
Abstract:
|
A major objective of longitudinal studies is to evaluate the dynamic patterns of the conditional distribution functions for multivariate outcomes over time. Existing methods for longitudinal analysis focus on modeling the conditional means, correlations and distributions separately using different regression models. These approaches lack cohesiveness when the scientific objective requires a unified model that simultaneously describes the conditional means, correlations and distributions. We develop a class of nonparametric dynamic copula models that incorporates the time-varying means, covariances and distributions into a unified regression structure. In this approach, we assume that the conditional distributions of the outcome variables belong to some known copula families when the time points are fixed, while the copula parameters are dynamic functions of time. We propose a general basis approximation method to estimate the dynamic parameters and distribution functions, and demonstrate its appropriate statistical properties through an epidemiological study of childhood cardiovascular risks and a simulation study. Theoretical justification for the basis approximation method is dem
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2019 program
|