Explanation behind proc mixed results

[teli4a]

Dear all,

It's my first post here, so hopefully I've not messed up something.

I've used the following code to analyze a data set:

proc mixed data=datata method=ml ;
class subject sex Form;
model time = sex|form|age@2 / s;
repeated / type=un subject=subject r;
lsmeans form sex ;
lsmestimate form "C vs S" -1 1 ;
lsmestimate sex "F vs M" -1 1 ;

run;

Can someone explain why in the solution for fixed effects I get Estimates equal to 0 for some effects and no info for their standard error, DF, t Value etc.

Thanks in advance

[ChrisNZ]

Moved thread to: Procedures

[sld]

We would need more information to answer your question. See the posting guidelines here:

https://communities.sas.com/t5/Getting-Started/How-to-get-fast-helpful-answers/ta-p/226133/jump-to/f...

In particular, provide a description of your experimental design, code, log, output, and your data or an example dataset that duplicates your problem.

The zero values for certain estimates could be OK (and are a consequence of the default coding system used in the MIXED procedure) or could indicate a problem with the model.

[PaigeMiller]

You get some zero estimates because there aren't enough degrees of freedom; in other words, if you have a class variable with 4 levels, you can only estimate the coefficient for three of them, and the fourth is set to zero and it's variability cannot be estimated so the variability is set to missing.

You should be looking at the least squares means to help you understand the model, instead of the model coefficients. The least squares means do not get set to zero, and so are much more interpretable.