Overview: HPMIXED Procedure

The experimental HPMIXED procedure uses a number of specialized high-performance techniques to fit linear mixed models with variance component structure. The HPMIXED procedure is specifically designed to cope with estimation problems involving a large number of fixed effects, a large number of random effects, or a large number of observations.

The HPMIXED procedure complements the MIXED procedure and other SAS/STAT procedures for mixed modeling. On the one hand, the models supported by the HPMIXED procedure are a subset of the models that you can fit with the MIXED procedure, and the confirmatory inferences available in the HPMIXED procedure are also a subset of the general analyses available with the MIXED procedure. On the other hand, the HPMIXED procedure can have considerably better performance than other SAS/STAT mixed modeling tools, in terms of memory requirements and computational speed.

A mixed model can be large in a number of ways, not all of which are suited for the specialized algorithms and storage techniques implemented in the HPMIXED procedure. The following are examples of linear mixed modeling problems for which the HPMIXED procedure has been specifically designed:

• linear mixed models with thousands of levels for the fixed and/or random effects

• linear mixed models with hierarchically nested fixed and/or random effects, possibly with hundreds or thousands of levels at each level of the hierarchy