Hi, I am comparing 3 treatments on a continuous measure administered at 3 time points (1, 2, 3) via a mixed model using the following code: proc mixed data=dataset METHOD=ML COVTEST EMPIRICAL;
class RECORD_ID time(ref='1') treatment(ref='CT');
model ISI_Total = TREATMENT TIME age female TREATMENT*TIME / SOLUTION DDFM=BW;
RANDOM intercept / TYPE=UN SUBJECT = record_id(TREATMENT);
SLICE TREATMENT*TIME / SLICEBY TIME;
LSMEANS TIME TREATMENT*TIME / DIFF ;
TITLE1 "Mixed Model: ISI_Total predicted by fixed Treat, Time and Treatment*Time, random intercept";
TITLE2 "CT is the reference treatment";
run; The SLICE statement generates an F-test for TREATMENT within each TIME that I need, but I also need to generate an F-test for change between time points (change from TIME=1 to TIME=2, TIME=1 to TIME=3, TIME=2 to TIME=3) by treatment within the same mixed model. I am having a hard time modifying my code to generate these F-tests within the SLICE statement. I would appreciate any guidance.
... View more
