PROC MIXED not computing any p-values

[alexandra-vita]

Hi,

I am using PROC MIXED with the code below, but the results are not displaying any p-values in both the Solution for Fixed Effects table and the Type 3 Tests of Fixed Effects table. In place of a p-value, there is a "." 

Is there something missing from the code that is preventing and/or interferring with SAS computing the p-values?

PROC MIXED
DATA= GGGImp.TNFa METHOD= REML;
CLASS sequence (ref= "AB") treatment (ref="A") time (ref="V1") period (ref= "P1");
MODEL TNFalpha = sequence treatment time period / SOLUTION;
RANDOM intercept / SUBJECT=subjID g vcorr;

RUN;

Thanks!

[JOL]

It sounds like there might be an issue with the degrees of freedom method or the structure of your model. Here are a few things you can check and try:

Degrees of Freedom Method: Sometimes, the default method for calculating degrees of freedom can cause issues. You can try specifying a different method, such as ddfm=KR (Kenward-Roger) or ddfm=BW (Between-Within), in your MODEL statement:

MODEL TNFalpha = sequence treatment time period / SOLUTION ddfm=KR;

[PaigeMiller]

@alexandra-vita 

It would be extremely helpful if you show us the output from that code so we can see what you are seeing. Create a screen capture of the output and include it in your reply using the "Insert Photos" icon; do NOT attach files.

I agree that it is likely you do not have enough degrees of freedom in your data, and this happens because you have as many terms in the model as data points. But SHOW US the output.