glmmplus: An R Package for Messy Longitudinal Data

• Activity Number: 397
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section for Statistical Programmers and Analysts
• Author(s): Ben Ogorek*+ and Caitlin Hogan
• Companies: Google and Google
• Keywords: imputation ; random effects ; R ; longitudinal ; variable selection ; missing data

Abstract

In modeling tasks involving large longitudinal data sets, there is often the need for random effects, grouped predictor terms, missing data forgiveness, nonlinear link functions, and variable selection capabilities. Many existing R packages focus on one of these problems, but the separate sets of functionality do not always integrate seamlessly. The glmmplus package addresses this problem by offering a wrapper to trusted packages such as mice and lme4, and adding new functionality such as Fast False Selection Rate (FSR) control for both forward and backward selection. The result is a la carte functionality to the user for messy longitudinal data. An analysis is presented from the National Longitudinal Survey.