proc mixed (repeated measurement and unstructered variance matrix)

[sousimou]

Hi everyone,

my data is regarding differents measurments over time for subjects.
I am using proc mixed to model data with repeated measurment, with
unstructered var-covar matrix. I want to extract the variance between
and within subjects for each group and at each time. Can you help me
how to do it?

this is my code :

%LET sub_var=.5;
%LET res_var=.9;
%LET nsb=50;
%LET seed=1928;
%LET nb_grp=3;

DATA tme;
Do i=1 to &nb_grp.;
DO subj = 1 to &nsb.;
DO time=1 to 4;
y = &sub_var.*rannor(&&seed.*i) + 1*(time=1)+ 2*(time=2)+
2*(time=3)+3*(time=4)
+ &res_var. *rannor(&&seed.*i);
group=compress("group"||i);
output;
end;
end;
END;
RUN;

PROC MIXED DATA=tme METHOD=ML NOCLPRINT=10 SCORING=3 CL;
CLASS time group subj;
MODEL y = time group group*time
/ DDFM=KR OUTPRED=predict RESIDUALS;
REPEATED time / TYPE=UNR SUBJECT=subj GROUP=group /*rcorr*/ R;
ODS OUTPUT COVPARMS=cov;
RUN;

thank you