Abstract for Part I:

We use some examples to give a very visual introduction to the main concepts of mixed models for longitudinal data analysis followed by a quick overview of SAS code and output illustrating the use of PROC MIXED, PROC GLIMMIX and PROC NLMIXED.   

Abstract for Part II:

What are some limitations of mixed models? When are estimated effects biased?  Must we then revert to fixed effects models? We will consider some solutions to problems in the use of mixed models including the role and interpretation of contextual variables.

Instructor:

Georges Monette is an Associate Professor in the Department of Mathematics and Statistics at York University. He received his Ph.D. in Statistics at the University of Toronto in 1980. Since 1985 he has participated in the Statistical Consulting Service at York University serving academic clients in a broad range of disciplines.  For the past ten years most of his work has involved hierarchical and longitudinal models, including graduate courses, workshops, seminars, collaboration and consulting.  In 2000, he organized the first Summer Programme in Data Analysis at York University which provided intensive training in hierarchical and longitudinal data analysis for researchers in Canada.