Date:         Fri, 4 Mar 2005 12:56:30 -0500
Reply-To:     Arthur Tabachneck <art297@NETSCAPE.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Arthur Tabachneck <art297@NETSCAPE.NET>
Subject:      Re: GLM and collinearity
Comments: To: Daniel Cournoyer <dcournoyer@SYMPATICO.CA>

Daniel,

I have to ask why you are concerned?  I believe you will get the same
result regardless of which level is used as a dummy.

Art
---------
On Fri, 4 Mar 2005 00:49:37 -0500, Daniel Cournoyer
<dcournoyer@SYMPATICO.CA> wrote:

>Hi to All,
>
>When using PROC GLM to perform GLM (!) for a model where a class variable
is
>used as predictor, and where we know that one level will be dropped due to
>collinearity, is it possible to choose which level will be dropped by SAS?
>By default, the last one is dropped; in R, it is the first one that is
>dropped. Can I tell SAS which one I would want it to drop without having
to
>go through the manual way, i.e. create indicator variables and not include
>the one I want dropped.
>
>Thank you,
>
>D