• support.sas.com

• Knowledge Base

• Support

• Training & Bookstore

• Happenings

• Support Communities

Mixed Model Procedures

• GLIMMIX — Generalized linear mixed models
• HPMIXED — Linear mixed models with simple covariance component structures by sparse-matrix techniques
• MIXED — General linear models with fixed and random effects
• NLMIXED — Mixed models in which the fixed or random effects enter nonlinearly

Mixed Models

A mixed model is a model that contains fixed and random effects. Over the last few decades virtually every form of classical statistical model has been enhanced to accommodate random effecs. The linear model has been extended to the linear mixed model, generalized linear models have been extended to generalized linear mixed models, and so on. In parallel with this trend, SAS/STAT software offers a number of classical and contemporary mixed modeling tools.

Below are highlights of the capabilities of the SAS/STAT procedures that fit mixed models:

• linear mixed models
  • fixed and random effects
  • REML, maximum likelihood, and MIVQUE0 estimation methods
  • least-squares means and differences
  • sampling-based Bayesian analysis
  • many covariance structures, some of which are compound symmetry, unstructured, AR(1), Toeplitz, heterogeneous AR(1), and Huynh-Feldt
  • multiple comparison of least-squares means
  • repeated measurements analysis
• generalized linear mixed models
  • conditional on normally distributed random effects, the data can have any distribution in the exponential family
  • covariance structures are modeled parametrically
  • flexible covariance structures for random and residual random effects, including variance components, unstructured, autoregressive, and spatial structures
  • produce hypothesis tests and estimable linear combinations of effects
  • mixed model smoothing and mixed model splines
  • joint modeling of heterocatanomic multivariate data
• nonlinear mixed models
• nested models

Statistics and Operations Research Home Page | SAS/STAT Software