Date:         Wed, 22 Oct 2008 15:20:45 -0400
Reply-To:     Peter Flom <peterflomconsulting@mindspring.com>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Peter Flom <peterflomconsulting@MINDSPRING.COM>
Subject:      Re: output r-square from proc logistic
Comments: To: haoyugu@GMAIL.COM
Content-Type: text/plain; charset=UTF-8

haoyugu@GMAIL.COM wrote

>Anyone know how to output rsquare into data "test1" in the following
>proc logistic?
>
>proc logistic data=aaa outest=test1;
>class M_Home  ;
>model kk = M_Home R_Home /Rsquare;
>quit;
>
>"test1" does not have rsquare, and it seems I can't use 'rsquare'
>option as in proc reg:
>proc logistic data=working.exp_keystatistics outest=test1 rsquare;
>


There is no exact equivalent to R squared for logistic regression, although there are a bunch of measures that are something like it.

see, e.g. Hosmer and Lemeshow, Applied Logistic Regression p. 164-166

In particular, they note that all the psuedo R squared measures are much lower in logistic than in OLS regression, even when the model is good.

Peter





Peter L. Flom, PhD
Statistical Consultant
www DOT peterflom DOT com