Comparison of Weighting Approaches for Longitudinal Data with Time-Dependent Cluster Sizes

• Activity Number: 499
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Matthew Stephenson*+ and Ayesha Ali and Gerarda Darlington
• Companies: University of Guelph and University of Guelph and University of Guelph
• Keywords: correlated outcomes ; informative cluster size ; cluster weights ; estimating equations

Abstract

Generalized estimating equations allow for modeling of correlated data provided that any missing data are missing completely at random (MCAR). However, there are many instances in which the MCAR assumption is violated, such as the presence of informative cluster sizes. It has been shown that cluster weighted generalized estimating equations allow for valid parameter estimation in the presence of informative cluster sizes and can be used in a longitudinal setting when cluster sizes remain fixed over time. Here we consider the setting in which cluster sizes may change over time. Through Monte Carlo simulation, we compare the performance of several weighting schemes, including time-dependent weights.