|
|
|
This is the preliminary program for the 2007 Joint Statistical
Meetings in Salt Lake City, Utah.
|
|
|
The views expressed here are those of the individual authors and not necessarily those of the ASA or its board, officers, or staff. Back to main JSM 2007 Program page |
= Applied Session,
= Theme Session,
= Presenter| CE_09C | Sun, 7/29/07, 8:30 AM - 5:00 PM | CC-150 G |
| Generalized Linear Mixed Models: Theory and Applications - Continuing Education - Course | ||
|
ASA |
||
| Instructor(s): Oliver Schabenberger, SAS Institute Inc. | ||
| The workshop discusses the theory of generalized linear mixed models and the application of these models. The course consists of two parts. Part I covers the requisite theory of generalized linear mixed models. Part II covers examples and applications and adds additional theoretical detail as needed; it is the majority of the course. The first part of the workshop makes the connection between linear models (LM), generalized linear models (GLM), linear mixed models (LMM), and generalized linear mixed models (GLMM) in terms of model formulation, distributional properties, and approaches to estimation. Participants learn that GLMMs are an encompassing family of models and come to understand the differences and similarities in approaches to estimation and inference within the class. The workshop discusses the pros and cons of various estimation approaches and describes their implementation with SAS/STAT® software. The first part of the course ends with a discussion of over-arching issues the analyst must confront when working with correlated, non-normal data, such as managing overdispersion and using marginal versus conditional models. The second part of the workshop uses a variety of examples to revisit the theory taught in Part I, to develop new insights (e.g., low-rank mixed model smoothing), and to present applications from different disciplines. The applications range from modeling rates, proportions and counts with random effects, to GEE-type marginal models for non-normal data, to mixed model smoothing. A final section describes inferential procedures following parameter estimation, for example, estimation of complex linear hypotheses, multiplicity adjustments, and adjusted standard errors. Computations are based on the mixed model tools in SAS/STAT software, primarily the GLIMMIX procedure and the NLMIXED procedure. | ||
|
JSM 2007
For information, contact jsm@amstat.org
or phone (888) 231-3473. If you have questions about the Continuing Education program,
please contact the Education Department. |