|
| From: | Jason Stover |
| Subject: | Re: GLM and interactions |
| Date: | Fri, 8 Jul 2011 19:03:01 -0400 |
| User-agent: | Mutt/1.5.18 (2008-05-17) |
I wonder if, in the presence of interactions, it would be easier to just encode the means for each factor/level combination, rather than encoding products of explanatory variables. -Jason
| [Prev in Thread] | Current Thread | [Next in Thread] |