I'm fitting a 3 level mixed model and trying to compute empirical rather than model-based standard errors for the fixed effects. When specifying the empirical option in PROC MIXED, I get a warning that I only have one effective subject. There are a lot of subjects in the data set, so i'm unsure what the problem is. Here is my code.
PROC MIXED DATA = EngScores METHOD = ML empirical; CLASS id t region sex; MODEL achieveEng = edu age sex time time*time / solution; REPEATED t/type=un subject=id; RANDOM intercept / subject=region; run;