The GLIMMIX Procedure

Basic Features

The GLIMMIX procedure enables you to specify a generalized linear mixed model and to perform confirmatory inference in such models. The syntax is similar to that of the MIXED procedure and includes CLASS , MODEL , and RANDOM statements. For instructions on how to specify PROC MIXED REPEATED effects with PROC GLIMMIX, see the section Comparing the GLIMMIX and MIXED Procedures. The following are some of the basic features of PROC GLIMMIX.

  • SUBJECT= and GROUP= options, which enable blocking of variance matrices and parameter heterogeneity

  • choice of linearization approach or integral approximation by quadrature or Laplace method for mixed models with nonlinear random effects or nonnormal distribution

  • choice of linearization about expected values or expansion about current solutions of best linear unbiased predictors

  • flexible covariance structures for random and residual random effects, including variance components, unstructured, autoregressive, and spatial structures

  • CONTRAST , ESTIMATE , LSMEANS , and LSMESTIMATE statements, which produce hypothesis tests and estimable linear combinations of effects

  • NLOPTIONS statement, which enables you to exercise control over the numerical optimization. You can choose techniques, update methods, line search algorithms, convergence criteria, and more. Or, you can choose the default optimization strategies selected for the particular class of model you are fitting.

  • computed variables with SAS programming statements inside of PROC GLIMMIX (except for variables listed in the CLASS statement). These computed variables can appear in the MODEL , RANDOM , WEIGHT , or FREQ statement.

  • grouped data analysis

  • user-specified link and variance functions

  • choice of model-based variance-covariance estimators for the fixed effects or empirical (sandwich) estimators to make analysis robust against misspecification of the covariance structure and to adjust for small-sample bias

  • joint modeling for multivariate data. For example, you can model binary and normal responses from a subject jointly and use random effects to relate (fuse) the two outcomes.

  • multinomial models for ordinal and nominal outcomes

  • univariate and multivariate low-rank mixed model smoothing