0

Is there a way in SAS to compare two regression models using ANOVA. What i want to replicate is - in R if i have 2 models - model1 & model2 i can directly run anova(model1, model2) to find if there is a significant difference between the two.

Is there a way to do the same in SAS.

merv
  • 67,214
  • 13
  • 180
  • 245
Raj
  • 1,049
  • 3
  • 16
  • 30

1 Answers1

1

No, because SAS doesn't store models that way. However, you can run each model (in PROC GLM or whatever) and then compare the results. You can get some of this by looking at the different "types" of error, too.

Peter Flom
  • 2,008
  • 4
  • 22
  • 35
  • 1
    Thanks. Guess it would be easier if SAS does incorporate some of the features available in R in the next version. – Raj Mar 02 '16 at 14:40
  • Adding a feature like this would be a radical change for SAS. It would probably require a whole redesign. I don't see it happening. R is (at least somewhat) object oriented. SAS is not. – Peter Flom Mar 02 '16 at 21:58