| Date: | Sat, 28 Apr 2007 10:58:29 -0700 |
|---|
| Reply-To: | Heferbob <heathandchunfang.Yates@GMAIL.COM> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Heferbob <heathandchunfang.Yates@GMAIL.COM> |
|---|
| Organization: | http://groups.google.com |
|---|
| Subject: | Re: Ridge Regression and PRESS statistic (Proc Reg Ridge + PRESS?) |
|---|
|
| In-Reply-To: | <1177643962.977768.291150@t38g2000prd.googlegroups.com> |
|---|
| Content-Type: | text/plain; charset="iso-8859-1" |
|---|
I thank all of those who have responded to me by email. I think I can
clear up alot of confusion about what I need. I really need to run a
Ridge Regression and get PRESS values for each k statistic. This is
pretty urgent as I need to figure out how to duplicate an example in
order to do assigned work.
The output in the example has this:
k (Parameter values go here) PRESS
0.00 ...
32,195,222
... ... ....
0.23 ...
15,584,595
I can get the PRESS value 32, 195, 222 by using the code I gave above
when I first wrote this problem. This is because PRESS is dumped into
the Outest data set, but only for the parameters, not the Ridge
Regression. That means it leaves PRESS value at k>0.00 blank. So I
think I have to start from scratch? I am fairly new to SAS, so I need
help. How can I produce a Ridgre Regression and PRESS?
Thanks,
Heath
|
