The PLM Procedure

Overview: PLM Procedure

The PLM procedure performs postfitting statistical analyses for the contents of a SAS item store that was previously created with the STORE statement in some other SAS/STAT procedure. An item store is a special SAS-defined binary file format used to store and restore information with a hierarchical structure.

The statements available in the PLM procedure are designed to reveal the contents of the source item store via the Output Delivery System (ODS) and to perform postfitting tasks such as the following:

  • testing hypotheses

  • computing confidence intervals

  • producing prediction plots

  • scoring a new data set

The use of item stores and PROC PLM enables you to separate common postprocessing tasks, such as testing for treatment differences and predicting new observations under a fitted model, from the process of model building and fitting. A numerically expensive model fitting technique can be applied once to produce a source item store. The PLM procedure can then be called multiple times and the results of the fitted model analyzed without incurring the model fitting expenditure again.

The PLM procedure offers the most advanced postprocessing techniques available in SAS/STAT software. These techniques include step-down multiplicity adjustments for p-values, F tests with order restrictions, analysis of means (ANOM), and sampling-based linear inference based on Bayes posterior estimates.

The following procedures support the STORE statement for the generation of item stores that can be processed with the PLM procedure: GENMOD, GLIMMIX, GLM, GLMSELECT, LIFEREG, LOGISTIC, MIXED, ORTHOREG, PHREG, PROBIT, SURVEYLOGISTIC, SURVEYPHREG, and SURVEYREG. The RELIABILITY procedure in SAS/QC software also supports the STORE statement. For details about the STORE statement, see the section STORE Statement of Chapter 19: Shared Concepts and Topics.