BookmarkSubscribeRSS Feed
ProfB
Fluorite | Level 6

I'm trying to pool type 3 tests from a regression model using proc genmod, across 10 imputed data sets. Is there a macro to leverage the ods type3 output table I can use to generate a pooled estimate of the p-value (for example, for "X1" in the code snippet below)?

 

 

proc genmod data=data plots=(dfbetacs); 
class id time;
model y=time X1 time*X1 X2*time X3*time /noint dist=normal link=identity id=studyid type3;
repeated subject=id/type=ind ecovb;
weight &weight;
ods output GEEEmpPEst=emp_est type3=type3table; ;
by _imputation_;
run;

proc mianalyze parms (classvar=level)= emp_est;
class time ;
modeleffects time*X1;
run;

 

2 REPLIES 2
SAS_Rob
SAS Employee
While there is no way to combine the Type3 tests in MIANALYZE you can combine the Chi-Square statistics themselves from the Type3 tests using the COMBCHI macro found on Paul Allison's web site.
http://www.ssc.upenn.edu/~allison/


ProfB
Fluorite | Level 6

Thanks, I will give this a shot!

SAS Innovate 2025: Call for Content

Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 16. Read more here about why you should contribute and what is in it for you!

Submit your idea!

What is ANOVA?

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.

Discussion stats
  • 2 replies
  • 2650 views
  • 1 like
  • 2 in conversation