Growing Up Fast: SAS® 9.2 Enhancements to the GLIMMIX Procedure
Oliver Schabenberger, SAS Institute, 2007
Enhancements fall into several categories: statistical estimation, model building, post-processing, and miscellaneous other features. For example, new estimation methods and covariance structures enable parameter estimation with reduced bias in more flexible generalized linear mixed models. A new statement provides a comprehensive facility to test hypotheses about covariance parameters. New multiplicity corrections offer more ways to adjust inferences.
The goal of this paper is to highlight the new estimation and inference techniques and to demonstrate their usage with examples.