Variable Selection for Generalized Linear Mixed Models by L1-Penalized Estimation

Groll, Andreas and Tutz, Gerhard (3. June 2011): Variable Selection for Generalized Linear Mixed Models by L1-Penalized Estimation. Department of Statistics: Technical Reports, No.108

Abstract

Generalized linear mixed models are a widely used tool for modeling longitudinal data. However, their use is typically restricted to few covariates, because the presence of many predictors yields unstable estimates. The presented approach to the fitting of generalized linear mixed models includes an L1-penalty term that enforces variable selection and shrinkage simultaneously. A gradient ascent algorithm is proposed that allows to maximize the penalized loglikelihood yielding models with reduced complexity. In contrast to common procedures it can be used in high-dimensional settings where a large number of otentially influential explanatory variables is available. The method is investigated in simulation studies and illustrated by use of real data sets.

• Item Type: Paper
• Keywords: Generalized linear mixed model, Lasso, Gradient ascent, Penalty, Linear models, Variable selection
• Faculties: Mathematics, Computer Science and Statistics > Statistics > Technical Reports; Mathematics, Computer Science and Statistics > Mathematics > Workgroup Financial Mathematics
• Subjects: 500 Science > 500 Science; 500 Science > 510 Mathematics
• Language: English