Professor Nan Lin

Department of Mathematics, Washington University

Title: Bayesian analysis of longitudinal data with mixed models and the robustness issue

Abstract: Linear mixed effects models are frequently used to analyze longitudinal data due to their flexibility in modeling the covariance structure and convenience to cope with missing data. We will illustrate how to perform Bayesian inference in linear mixed effect models. Moreover, a normal distribution is often assumed for the random effects and residuals, and  this makes inferences vulnerable to the presence of outliers. We will also discuss how to obtain robust inference by using heavy-tailed distributions.