topic Re: How to predict the fixed effects and fixed & random effects when using PROC MIXED? in Statistical Procedures

How to predict the fixed effects and fixed & random effects when using PROC MIXED?

CynicalFire — Sat, 07 Dec 2019 22:21:38 GMT

I am using PROC MIXED in SAS Studio. When looking through communities here, I found this one response that sort of answers my question, but the problem is, I don't understand how to use it.

The OUTM= option on the model statement gives you just the predicted fixed-effects means (and associated statistics, such as standard errors). The OUTP= option gives you the predictions for fixed and random effects (i.e., the EBLUPS), with associated statistics.

The following is my current code, and I don't quite understand how to put the OUTM and the OUTP options in my model.

PROC MIXED data=bmi METHOD=ML COVTEST;
CLASS studyid;
MODEL ChildhoodBMIz = age sex age*age /SOLUTION;
RANDOM INTERCEPT age /SUB=studyid TYPE=UN G GCORR;
RUN;

However, if there is another, easier way to do accomplish what I am trying to do, then I will gladly do that instead.

Re: How to predict the fixed effects and fixed & random effects when using PROC MIXED?

PaigeMiller — Sat, 07 Dec 2019 22:47:53 GMT

The OUTP= option and the OUTPM= option (no such option as OUTM=) go in the MODEL statement.

https://documentation.sas.com/?docsetId=statug&docsetTarget=statug_mixed_syntax10.htm&docsetVersion=14.3&locale=en#statug.mixed.modelstmt_outpm

Re: How to predict the fixed effects and fixed & random effects when using PROC MIXED?

CynicalFire — Sat, 07 Dec 2019 23:59:52 GMT

Okay, thank you for the link, but I still do not understand how to put it into my code.

Re: How to predict the fixed effects and fixed & random effects when using PROC MIXED?

PaigeMiller — Sun, 08 Dec 2019 00:02:16 GMT

MODEL ChildhoodBMIz = age sex age*age /SOLUTION outp=sas_data_set_name outpm=sas_data_set_name2;