| Date: | Thu, 26 Oct 2006 14:44:48 -0400 |
|---|
| Reply-To: | Peter Flom <Flom@NDRI.ORG> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Peter Flom <Flom@NDRI.ORG> |
|---|
| Subject: | Re: Stepwise Regression |
|---|
|
| In-Reply-To: | <1161887581.345658.153960@h48g2000cwc.googlegroups.com> |
|---|
| Content-Type: | text/plain; charset=US-ASCII |
|---|
>>> Vadim Pliner <Vadim.Pliner@VERIZONWIRELESS.COM> 10/26/2006 2:33 pm >>> wrote
<<<
b. I was not talking specifically about linear stepwise regression.
AFAIK, Partial Least Squares is not applicable to the case when
dependent variable is binary. I know there are alternatives to stepwise
logistic regression for selecting variables that you might consider
"better" as well, but see a. above.
>>>
Maybe classification trees and the various sophisticated add-ons like bagging and forests and so on.
Or, if it's pure prediction, model averaging seems to offer a good deal
I am usually not in the 'pure prediction' mode, though
Peter
| 
