The National Program on
Complex Data Structures


at the University of Toronto, Medical Science Building
August 14 -17, 2006 -- 9 AM to 5 PM

Director: Paul N Corey
Co-directors: Jamie Stafford and Wendy Lou
Lecture 3: GENERALIZED MIXED MODELS -- Joseph Beyene

Generalized linear mixed models (GLMMs) are a generalization of linear mixed models and can be used to fit data that may be assumed to follow distributions in the exponential family. Traditional regression methods, such as logistic and Poisson regression models, assume independent observations and are unsuited to analyzing data with complex structure. Mixed model regression approaches, such as GLMMs, on the other hand provide a framework for analyzing data with dependent observations and allow proper modeling of heterogeneity. With recent advances in statistical software that can be used to fit a variety of mixed models, practitioners and researchers alike are increasingly appreciating the utility of mixed models. This lecture will focus on generalized linear mixed models for discrete data. Emphasis will be on concepts, applications and interpretations over mathematical technical details. Marginal models versus conditional approaches will be compared and contrasted and illustrative examples along with sample SAS codes will be provided.

Joseph Beyene has a Doctorate from the University of Toronto in the field of Biostatistics and currently holds the positions of Scientist at the Hospital for Sick Children Research Institute, Toronto, Assistant Professor in the Department of Public Health Sciences and Department of Health Policy, Management & Evaluation at the University of Toronto, and Assistant Professor (part-time) in the Clinical Epidemiology & Biostatistics Department, McMaster University. He is interested in the development and application of statistical methods in a wide range of health research problems. His specific areas of statistical research interests include systematic reviews/meta-analysis, statistical genomics, and generalized linear models.

Back to Summer School Index