Statistical Methods for Dealing with Missing Data in Longitudinal Studies

• Activity Number: 337 - Approaches for Modeling Clustered and Longitudinal Data
• Type: Contributed
• Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Biometrics Section
• Author(s): Panpan Zhang* and Sharon Xiangwen Xie
• Companies: University of Pennsylvania and University of Pennsylvania
• Keywords: Longitudinal analysis; Missing data; Multiple imputation

Abstract

Missing data is common in longitudinal analysis, rendering it a burgeoning research area in the community. In this presentation, we discuss and compare prominent methods for handling missing data in longitudinal analysis, for instance, multiple imputation. More specifically, we evaluate the performances of the considered methods through bias and efficiency gains under different classes of model settings. We present our results via extensive simulation studies and a real data set related to neurodegenerative diseases.