A Comparison of One-Step and Three-Step Approaches in the Shared Parameter Growth Mixture Model

• In Event: 52.076 - Longitudinal Data Modeling

Abstract

The shared parameter growth mixture model has been proposed as a method to handle MNAR data in longitudinal studies where the probability of missing values is dependent on the individual underlying growth trajectory. This study aims to compare the one-step approach with a three-step approach for adding covariates in the shared parameter growth mixture model. Moreover, since the classes are related to the missing data patterns in shared parameter growth mixture models, determination of the class is important aspect for this model. Hence, this study evaluates whether the one-step or three-step approaches differ with respect to the placement of individuals into classes and estimation of class-specific parameters.

Authors

• Dee Duygu Cetin-Berber, University of Florida

• Walter L. Leite, University of Florida