[Coming soon]
proc glimmix data=train_data absconv=0.005;
model y = &covars /s;
random &z /s type=rsmooth knotmethod=equal(20);
run;
proc glimmix data=test nofit outdesign=test2;
model y=&covars /s;
random &z /s type=rsmooth knotmethod=equal(20);
run;
proc score data=test2 score=beta_fix type=parms out=score_fix;
var &covars;
run;
proc score data=test2 score=beta_random type=parms out=score_random;
var _z:;
run;
Reference:
1. SAS Institute, Statistical Analysis with the GLIMMIX procedure Course Notes, SAS Press, SAS Institute
2. D Rupper, M.P. Wand, R.J. Carroll, Semiparametric Regression, Cambridge University Press, Cambridge, 2003
5 comments:
How are beta_fixed and beta_random obtained? I get that they are the parameter estimates, but it seems they need to be in a special format. Any suggestions?
estimates of fixed effects can be obtained via ODS OUTPUT ParameterEstimates= ;
Empirical Bayesian Prediction can be obtained via ODS OUTPUT SolutionR= ;
You need to transpose both data sets on VAR estimate;
Great technique & thanks for the post! Applied to model averaged parameter estimates after transposing and it worked great.
Are you aware of documentation describing how SAS scores new data, specifically the random effect? It would be beneficial to understand how a new x,y coordinate, the knot data, and the parameter estimate for each knot interact to produce a score for the random effect.
PROC PLM supports scoring GLIMMIX models. Check the SAS Doc for PROC PLM for details. Cheers!
Post a Comment