|Date: ||Fri, 21 Apr 2006 07:43:33 -0400|
|Reply-To: ||Jonas Bilenas <jonas.bilenas@CHASE.COM>|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||Jonas Bilenas <jonas.bilenas@CHASE.COM>|
|Subject: ||Re: Interpreting Odds Ratio|
Stepwise! Oh No!
Why rank transform? Doing some non-parametric logistic? Pure rank or did
you assign into deciles or other groups? How did you handle ties? Did you
have to worry about sampling?
On Thu, 20 Apr 2006 11:50:09 -0700, Thera <trtyner@GMAIL.COM> wrote:
>I have developed a binary logistic model with the stepwise option using
>rank transformed variables as well as log transformed variables. How
>do I interpret the odds ratio for these variables.