The TEST or MTEST statement in PROC REG does not support hypothesis testing across multiple MODEL statements. However, you can do this testing by using the STEST statement in SAS/ETS PROC SYSLIN. For example, the following statements test the hypothesis that the X1 and X2 parameters are equal in these two models that are fit by using ordinary least squares:

   proc syslin ols;
      model y1=x1;
      model y2=x2;
      stest y1.x1 = y2.x2;
      run;

This method can be extended to any number of models and can be used with other estimation methods that are supported by PROC SYSLIN (except FIML).

If you want to test the equality of parameters in models that are fit to independent groups, where the models use the same predictors, this can be done using an analysis of covariance (ANCOVA) approach as discussed here.

Operating System and Release Information

Product Family	Product	System	SAS Release
			Reported	Fixed*
SAS System	SAS/STAT	All	n/a
SAS System	SAS/ETS	All	n/a

* For software releases that are not yet generally available, the Fixed Release is the software release in which the problem is planned to be fixed.

---

Type:	Usage Note
Priority:	low
Topic:	SAS Reference ==> Procedures ==> REG Analytics ==> Multivariate Analysis SAS Reference ==> Procedures ==> SYSLIN Analytics ==> Econometrics Analytics ==> Structural Equations Analytics ==> Regression

Date Modified:	2005-07-01 07:35:28
Date Created:	2005-05-31 16:29:57