Bayesian Modeling of Between- and Within-Subject Variances Using Mixed Effects Location Scale Models for Intensive Longitudinal Data

• Activity Number: 413
• Type: Topic Contributed
• Date/Time: Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Donald Hedeker* and Robin J. Mermelstein and Xiaolei Lin
• Companies: The University of Chicago and University of Illinois at Chicago and The University of Chicago
• Keywords: variance ; mixed-effects model ; location-scale ; heterogeneity

Abstract

For longitudinal data, mixed models include random subject effects to indicate how subjects influence their responses. The error variance and the random-effects variance are usually considered to be homogeneous. These variances characterize the within-subjects and between-subjects variation in the data. In studies using Ecological Momentary Assessment (EMA) or other types of intensive longitudinal data collection, up to 30-40 observations are often obtained for each subject. We focus on an adolescent smoking study using EMA, where interest is on characterizing changes in mood variation associated with smoking. We describe, using Bayesian modeling, how covariates can influence the mood variances, and also describe an extension of the standard mixed model by adding a subject-level random effect to the within-subject variance. This permits subjects to have influence on the mean, or location, and variability, or scale, of their mood responses. We also allow the location and scale random effects to be correlated. These mixed-effects location scale models have useful applications in research areas where there is interest on the joint modeling of the mean and variance structure.