Date:         Thu, 22 Nov 2007 15:00:17 -0500
Reply-To:     Wensui Liu <liuwensui@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Wensui Liu <liuwensui@GMAIL.COM>
Subject:      Re: how to build a saturated model for logistic regression?
Comments: To: Peter Flom <peterflomconsulting@mindspring.com>
In-Reply-To:  <6242068.1195760979906.JavaMail.root@mswamui-chipeau.atl.sa.earthlink.net>
Content-Type: text/plain; charset=ISO-8859-1

thx, peter, as always,
i am more curious how to do it in sas.

On Nov 22, 2007 2:49 PM, Peter Flom <peterflomconsulting@mindspring.com> wrote:
> A saturated model is one with 0 degrees of freedom.
>
> One way would be to supply a 'subject' effect - that is, an effect for the individual person, or whatever the subject might be.
>
> Peter
>
>
> -----Original Message-----
> >From: Wensui Liu <liuwensui@GMAIL.COM>
> >Sent: Nov 22, 2007 2:36 PM
> >To: SAS-L@LISTSERV.UGA.EDU
> >Subject: how to build a saturated model for logistic regression?
> >
> >Dear Listers,
> >Sorry for bothering you on Thxgiving.
> >I am just curious how to build a saturated model for logistic
> >regression or other kinds of regression.
> >
> >Thank you so much!
> >
> >wensui
>
>



--
===============================
WenSui Liu
Statistical Project Manager
ChoicePoint Precision Marketing
(http://spaces.msn.com/statcompute/blog)
===============================