Obtaining Residuals in a Mixed Model and Using them as an Outcome in another analysis

[lgutm18]

Hello, 

I need to obtain the residuals and then use them as the outcome in a separate model. In the past, I was able to extract the residuals using proc reg, as shown below: 

proc reg data=abstract.pedometer_speedquint2;
model stepsperday=stepmin_avg_valid ;
OUTPUT out=b r=res;
run;

proc print data=abstract.pedometer_speedquint2;
var r; /*residuals per participant*/
run;

However, I have not been able to figure out how to do the same using proc mixed.

This is the code I am using for the analysis from which I want to extract the residuals: 

proc mixed data=trial4;
class idno time; 
model CASI= time WMFA WMH ICV TOTAL_GM_VOL TOTAL_WM_VOL AGE RACE GENDER/s chisq;
repeated time/type=un subject=idno r rcorr;
run;

Thanks!

[ballardw · Accepted solution]

To add residuals to an output statement use one or both of the OutPM=<some data set name> or Outp=<some other dataset name> along with the RESIDUAL option to the MODEL statment. OutPM gives the marginal residuals and OutP gives the conditional residuals.

proc mixed data=trial4;
class idno time; 
model CASI= time WMFA WMH ICV TOTAL_GM_VOL TOTAL_WM_VOL AGE RACE GENDER/s chisq  OUTPM=WORK.PM RESIDUAL;
repeated time/type=un subject=idno r rcorr;
run;

Read the DETAILS section of the Online help for Proc MIXED under "Residuals and Influence Statistics" for details.