Date:         Thu, 19 Apr 2001 09:41:02 -0700
Reply-To:     Dale McLerran <dmclerra@MY-DEJA.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Dale McLerran <dmclerra@MY-DEJA.COM>
Subject:      Re: Ridge Regression
Comments: To: cobas@asu.edu
Content-Type: text/plain

José,

I am not sure that this is the correct forum to really address this
particular question.  I have used ridge regression in the past, but
have not done so for quite a few years now, one reason being that
the data I work with mostly are group randomized trials data.  These
data require random effects regression models.  The ridge solution
for minimizing MSE is not available for the MIXED procedure.  For
that matter, I don't know if a ridge regression would even be
appropriate mathematically when using the MIXED procedure.  Really,
I think that you would have to do some literature search to see
whether anyone is using ridge regression these days.

Just out of curiosity, what motivated you to raise the issue here
on SAS-L?

Dale


>Date:         Thu, 19 Apr 2001 08:41:16 -0700
>Reply-To: "Jose A. Cobas" <cobas@ASU.EDU>
>From: "Jose A. Cobas" <cobas@ASU.EDU>
>Subject:      Ridge Regression
>To: SAS-L@LISTSERV.UGA.EDU
>
>Hi folks,
>What ever became of ridge regression? Is it not used anymore? Thanks in
>advance.
>
>
>
>José A. Cobas, Ph.D.
>Professor
>Department of Sociology
>Arizona State University
>Tempe, AZ 85287-2101
>Voice: (480) 965-3785
>Fax:    (480) 965-0064
>
>
>
><< msg2.html >>




---------------------------------------
Dale McLerran
Fred Hutchinson Cancer Research Center
mailto: dmclerra@fhcrc.org
Ph:  (206) 667-2926
Fax: (206) 667-5977
---------------------------------------

------------------------------------------------------------
--== Sent via Deja.com ==--
http://www.deja.com/