proc mixed data=MyBigDataSet; class Env A B sub; model y = A B x / ddfm=KenwardRoger; random int A*B / sub=Env; repeated / subject=Env*A*B type=AR(1); store sasuser.mixed; run; proc plm source=sasuser.mixed; show cov Parms; lsmeans A B / diff; score data=NewData out=ScoreResults; run;
The STORE statement in the PROC MIXED step requests that the MIXED procedure save those results that are needed to perform statistical tasks with the PLM procedure. For example, the MIXED procedure saves the necessary pieces of information that relate to the Kenward-Roger degree-of-freedom method. The results from the LSMEANS statement in the PROC PLM step thus apply this technique for calculating denominator degrees of freedom. The SHOW statement in the PLM procedure reveals the contents of the item store in terms of ODS tables, and the SCORE statement computes predicted values in a new data set. For more information about postprocessing tasks based on item stores, see the documentation for the PLM procedure.