https://communities.sas.com/t5/SAS-Procedures/How-to-run-regressions-with-more-than-one-class-variable-and/m-p/30523#M7256 Hi, I'm trying to run a regreession with two dummy variables, and I need to save the residuals and the estimators to be used later. I have tried using proc glm, but it doesn't work using more than one class variable.. Please help mee... Mon, 08 Nov 2010 21:04:53 GMT Multipledummyvariables 2010-11-08T21:04:53Z https://communities.sas.com/t5/SAS-Procedures/How-to-run-regressions-with-more-than-one-class-variable-and/m-p/30523#M7256 Hi, I'm trying to run a regreession with two dummy variables, and I need to save the residuals and the estimators to be used later. I have tried using proc glm, but it doesn't work using more than one class variable.. Please help mee... Mon, 08 Nov 2010 21:04:53 GMT https://communities.sas.com/t5/SAS-Procedures/How-to-run-regressions-with-more-than-one-class-variable-and/m-p/30523#M7256 Multipledummyvariables 2010-11-08T21:04:53Z https://communities.sas.com/t5/SAS-Procedures/How-to-run-regressions-with-more-than-one-class-variable-and/m-p/30524#M7257 hmmm<BR /> REG does not use the CLASS statement. So using dummy vars is an alternative. GLM does use the CLASS statement and does support more than one class variable. The OUTPUT statement is used to write out the individual estimates. Past that I am not sure of the question (or the model). Consider the following to see if it can start the discussion.<BR /> <BR /> [pre]proc glm data=sashelp.class;<BR /> class age sex;<BR /> model weight = age|sex height;<BR /> output out=stat p=pred cookd=cooksd;<BR /> run;<BR /> [/pre] Tue, 09 Nov 2010 03:11:33 GMT https://communities.sas.com/t5/SAS-Procedures/How-to-run-regressions-with-more-than-one-class-variable-and/m-p/30524#M7257 ArtC 2010-11-09T03:11:33Z