First of all,
Thank you everybody, and Richard in particular, for the very clear and comprehensive
explanation.
I agree, stepwise (as "all subsets" approach) is completely mindless and atheoretical...nevertheless
there are times (exploratory sudies) in which you do not have any clue about the potential predictors
and stepwise can be a good method to start.
I knew that backward approach is usually considered better than forward stepwise, just because
more conservative than that. The question is: since "mixed" stepwise integrate both the approaches,
should it be considered better than the simple backward ?
Cheers
Diego
----- Messaggio originale -----
Da: SR Millis <[email protected]>
A: [email protected]
Inviato: Sabato 27 gennaio 2007, 12:22:00
Oggetto: Re: st: "alternative" stepwise approaches: Hierachical and
More specifically, Harrell discusses the fast backward
elimination of variables based on the method of
Lawless and Singhal. It is implemented in Harrell's
fastbw function---which is available in his Design
library for R/Splus.
SR Millis
--- Joseph Coveney <[email protected]> wrote:
---------------------------
>
> The book isn't at-hand at the moment, but as I
> recall, Frank E. Harrell,
> Jr., in his _Regression Modeling Strategies_ (New
> York: Springer-Verlag,
> 2001) suggests backward selection if you're going to
> use stepwise procedures
> at all. As it turns out, this is the default in
> Stata.
>
> Joseph Coveney
Scott R Millis, PhD, MEd, ABPP (CN & RP)
Professor & Director of Research
Department of Physical Medicine & Rehabilitation
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email: [email protected]
Tel: 313-993-8085
~~~~~
This message, including any attachments, is intended for the person(s) or organization(s) named above. It is confidential and may be legally privileged or protected by law. Unauthorized use, copying, or disclosure of any part is strictly prohibited. If you are not the intended recipient, please contact the sender immediately and delete it from your system.
DISCLAIMER OF KNOWLEDGE OF VIRUS
Although this email and any attachments are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by Scott R Millis for any loss or damage arising in any way from its receipt or use.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
___________________________________
Vinci i biglietti per FIFA World Cup in Germania!
yahoo.it/concorso_messenger
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
