I am running PROC GLM for the fixed effects regression of variable &depVar. over independent variables &indVar., while controlling for fixed effects for the geographic variation HRR_x_Year. Both ways, the results must be equivalent; however, in first case it complains about a singularity and R-square jumps to 0.84, where as in later case the R-square due to HRR_x_Year only changes about 0.06.Is the implementation different intrinsically.
PROC GLM DATA=WORK.MODEL2;
absorb HRR_x_Year;
model &depVar.= &indVar. / SS3 SOLUTION ;
run;
"Note: The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable."
PROC GLM DATA=WORK.MODEL2;
Class HRR_x_Year;
model &depVar.= &indVar. HRR_x_Year / SS3 SOLUTION ;
run;
Is it possible that you have reversed the order of the two procedure calls?
The message
Note: | The X'X matrix has been found to be singular... |
is displayed when you use CLASS variables and you use the SOLUTION option to request the parameter estimates. The NOTE reminds you that the GLM parameterization of classification effects is a singular parameterization and therefore the last level of the classification variable will be set to zero.
Hi I know this is an old thread but I seem to be having the same problem, were you able to find a solution? If so do you mind sharing?
Please show us the different results.
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 25. Read more here about why you should contribute and what is in it for you!
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.