GEE for binary longitudinal data

[deleted_user]

Hi,
I am using proc genmod for binary longitudinal data. SAS code and error message listed below. Does this means I need exclude all the subjects which have missing measurements at any time point? Thanks for your help.

proc genmod data=ee descending;
class subjID sex time group;
model response_var=Sex group Weight /dist=bin link=logit;
repeated subject=subjID / withinsubject=time logor=exch;
run;

ERROR: There seems to be a problem in the WITHINSUBJECT effect data. Check to see that there are at least as many levels of the WITHINSUBJECT effect as there are measurements for each subject.

[deleted_user]

I was lucky because a search at support.sas.com pointed me to http://support.sas.com/kb/8/504.html