Simultaneous Bayesian Inference for Skew-Normal Mixed-Effects Joint Models for Longitudinal Data

• Activity Number: 521
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Yangxin Huang*+
• Companies: University of South Florida
• Keywords: Bayesian analysis ; covariate measurement errors ; HIV dynamics ; mixed-effects joint models ; skew-normal distribution

Abstract

Various mixed-effects models have been suggested for longitudinal data. Among those models are LME, NLME, and SNLME models. However, one often assumes that a model error is normally distributed, but this assumption may be unrealistic, particularly, if the data exhibit skewness. In addition, some covariates may be measured with substantial errors. This paper addresses these issues simultaneously by jointly modeling the response variable and a covariate process with measurement errors using a Bayesian approach to compare these three models. A real AIDS data set was used to illustrate the models and methods. It was found that there was a significant incongruity in the estimated decay rates based on the three mixed-effects models, suggesting that the decay rates by LME or NLME models should be interpreted differently from those estimated by SNLME models which is preferred to other models.