Date: Tue, 23 Nov 2004 11:40:36 -0600
Reply-To: Duck-Hye Yang <dyang@CHAPINHALL.ORG>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Duck-Hye Yang <dyang@CHAPINHALL.ORG>
Subject: Selecting referece category affects DF in hazard model???
Content-Type: text/plain; charset=US-ASCII
I got weird results when I changed reference category for hazard
I had four dummy variables for welfare programs: WIC only, Food Stamp
Only, both and neither.
When I used Both as reference, the results seem to be ok.
However, when I used Neither as reference, Food stamp only had no DF
and parameter estimate got zero.
Could you help me understand what is happening here?