Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 414 - Risk Modeling and Regression Techniques
Type: Contributed
Date/Time: Thursday, August 12, 2021 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #318800
Title: Joint Effect Selection in Generalized Linear Mixed Model Analysis
Author(s): Shou-En Lu* and Sinae Kim and Jerry Q. Cheng and Changfa Lin
Companies: Rutgers School of Public Health and Cancer Institute of New Jersey and Bristol Myers Squibb - Global Biopharmaceutical Company and New York Institute of Technology and Artech
Keywords: Generalized Linear Mixed Mode; selection of fixed and random effects; adaptive LASSO

Generalized linear mixed models (GLMMs) are commonly used to describe relationships between correlated responses and covariates. In this talk, we introduce a method to select both fixed and random effects in GLMMs simultaneously. Our proposed method is to construct the objective function using the confidence distribution based on the joint asymptotic distribution of the fixed effect and random effect parameter estimators for effect selection. With a proper choice of regularization parameters in the format of adaptive LASSO framework, we show the consistency and oracle properties of the proposed regularized estimators. Simulation studies have been done to assess the performance of the proposed estimators and demonstrate computational efficiency, compared to the existing selection procedures in GLMM setting. Our method has also been applied to a longitudinal breast cancer study to identify demographic and clinical factors associated with the incidence of common mammographic sequelae after breast conserving surgery and radiation therapy.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program