Hi there,
Sorry that I bothered you with the wrong observation
>Can anybody comment on the fact that
>
>summary(lm)$sigma .ne. sqrt(stepwise$rss/(length(data[1,])-stepwise$size))
>
>Explanation: I am using stepwise to assess various regressions, dim(a)=n say:
>
>sw <- stepwise( cbind(a,b,c,d), y,method=..).
>
>In my opinion the right hand side of the above unequality (e.g.
sqrt(sw$rss/(n-p)) should be a valid >estimate of the standard error for a
specified regression containing p variables of the a,b,c,d ; and >this
should have the same value as summary(regression with the same variables
a,b,c,d)$sigma.
>Or am I missing something? PS: no NAs are present in the data.
I misinterpreted my results, both sigma and sqrt are the same, as must be.
--Christian
Christian W. Hoffmann
Swiss Federal Institute for Forest, Snow and Landscape Research
CH-8903 Birmensdorf, Switzerland
phone: ++41-1-739 22 77 fax : ++41-1-739 22 15 e-mail: Hoffmann@WSL.CH
WWW: http://www.wsl.ch/land/dynamics/modeling/hoffmann.htm
FTP: ftp://ftp.wsl.ch/pub/hoffmann/
I prefer WWW pages which can be viewed best with *any* browser, contain
*no* cookies, *no* Java, *no* frames, and *no* required images !
*******************************************************
Do not miss the TOTAL ECLIPSE OF THE SUN on
1999 August 11, 12:35 MEST (at Stuttgart).
The center line will pass through Metz, Karlsruhe, Stuttgart, Muenchen,
into Austria up to Romania
duration of totality will be 2min 20sec. See the corona, protuberances !!
The southern limit of totality will pass through Offenburg, Lake Starnberg,
Berchtesgaden.
Zuerich will see the diameter of the sun being covered by 97.4%,
reducing sun light to about 1/60. See no corona, protuberances :-(
Information at:
http://www.astroinfo.org/eclipse/
********************************************************
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|