GLIMMIX: constant term only model

[Tissue5454]

Hi all!

I am constructing logistic regression model with random effiect for binary data by GLIMMIX.

At first, I am trying to test significance of  intercept random effect of cluster "fid".

To do this, I think it is convenient to  specify constant term only model with GLIMMIX.

Anyone know how I can specify constant term only model in GLIMMIX or any other appropriate way?

--------------------------------------------------------------------------------

proc glimmix data=logitbank method=quad;
class fid;
model ydat2(event='1') = ????
/dist = binary link=logit ddfm=none solution;
random intercept/ subject=fid;
output out =glmout pred =xbeta pred(ilink) =predprob;
covtest;
run;

---------------------------------------------------------------------------------

Thanks,

Tissuee5454

[Ksharp · Accepted solution]

You can leave it empty if you want only fit a intercept term.

model ydat2(event='1') = /dist = binary link=logit ddfm=none solution;

[Tissue5454]

HI Ksharp!

Thank you for your help!

I did not come up with that!

As you said, I could fit constant term only.

Thank you again!

Tissuee5454

[Ksharp]

I learned it from @Rick_SAS   's blogs .Search it you will find some rick's blogs about this topic .

[Tissue5454]

Hi Ksharp!

Thank you for letting me know!

I will check the blogs!

Tnahk you again,

Tissuee5454