Hi everyone,
Is it a problem if the subgroups specified by nesting in my RANDOM statement are unbalanced? For example, below you can see id(Diagnosis) where there are 5 diagnoses but the number of ids (patients) with each diagnosis range from 8 to 16 (e.g. 8 patients have diabetes, 16 patients have cancer, 10 patients have epilepsy). Do I need to adjust for this in my syntax? If so, what syntax do I need to add?
PROC GLIMMIX DATA=CONC_long;
CLASS Time (ref= '1') Diagnosis;
MODEL LoHiSev = Composite / dist=binomial link=logit solution;
RANDOM intercept / subject=id(Diagnosis) type=CS;
OUTPUT out=preAUCnested1 pred=xbeta pred(ilink)=predprob pred(ilink noblup)=fix_predprob;
RUN;
Thank you, SAS Community!
Best,
MeasuringHealth
Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.
Register today!ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.
Find more tutorials on the SAS Users YouTube channel.