multivariate multiple regression

#1
Hello all:

I'm working on a multivariate multiple regression analysis in SAS. I have 3 response variables and 2 independent variables.

When using the proc glm method in sas, I can do the the full factorial analysis (iv1, iv2, iv1 * iv2 interaction term) and I'm confident it is right. I confirmed my answer using SPSS.

however, when I try to fit do this same model using proc reg, I don't get similar results. I first created a new data step to define the interaction term. Then i used proc reg to model the 3 dvs on the 3 iv's. The results don't match those obtained from the proc glm method. I'm just wondering whats going on? Am I missing something? I can post my code and the original data if needed. Thanks