Longitudinal Data Analysis Using Generalized Nonlinear Mixed-Effects Modeling Framework

• In Event: Multilevel Models

Abstract

Longitudinal data are widely used in social and behavioral sciences to investigate the trajectory of the underlying phenomena under study over time. The analyses of abilities development (e.g., cognitive), evolution of a behavior, or treatment reaction over time are examples of settings where longitudinal data are needed. One intrinsic characteristic of longitudinal data is that repeated measures on the same individual are correlated. Several statistical models have been developed for different types of longitudinal data, however, a general framework can be defined where all the existing models are subsumed as special cases. The aim of this study is to present the generalized nonlinear mixed-effects models as a general class of models to analyze different types of longitudinal data.

Authors

• Yadira Peralta, University of Minnesota - Twin Cities

• Nidhi Kohli, University of Minnesota - Twin Cities

• Chun Wang, University of Minnesota