04-16-2014 11:47 PM
I am using proc glimmix to fit a multilevel logistical model. Den df in the parameter estimates of gender*TE*item is 0 so there is no p value. What might be the reason?
proc glimmix data=data1;
class tID sID item gender TE;
model response (Event='1')= item gender*TE*item / Dist=Binary link=logit solution noint DDFM=BW;
random intercept / subject=tID type=vc;
random intercept / subject=sID(tID) type=vc;
04-17-2014 09:34 AM
I am guessing you have unbalanced data, with a lot of cells with no observations (in the three-way table). I usually see this problem with the default ddf method (ddfm=containment). You could use ddfm=kr (for Kenward-Roger method). This will work for the default estimation method. However, the default estimation method is probably not the best choice for binary data because of bias of parameter estimates. I would suggest you use the option method=laplace on the procedure statement. But if you do that, you cannot use ddfm=kr. If you felt you had enough observations, you could use chi-squared wald statistics instead of F statistics for hypothesis testing and confidence intervals. Just put CHISQ as an option in the model statement.