home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky sci.math.num-analysis:3591 sci.math.stat:2604 sci.math.symbolic:3285
- Newsgroups: sci.math.num-analysis,sci.math.stat,sci.math.symbolic
- Path: sparky!uunet!rational.com!questor!davidm
- From: davidm@questor.Rational.COM (David Moore)
- Subject: Re: mathematica stepwise regression package
- Message-ID: <davidm.724451400@questor>
- Keywords: mathematica stepwise regression
- Sender: news@rational.com
- Organization: Rational
- References: <1446@ares.edsr.eds.com> <mcclella.724399758@yertle.Colorado.EDU> <1992Dec15.083258.1474@lth.se>
- Date: Tue, 15 Dec 1992 20:30:00 GMT
- Lines: 19
-
- andersh@maths.lth.se (Anders Holtsberg) writes:
-
-
- >gary mcclelland mcclella@yertle.colorado.edu:
-
- >You mean what is bad? Let's say we want to make predictions.
- >Is the idea of using a subset of the predictors that is bad or
- >do you mean the stepwise way to pick them? If the latter: do
- >you know any better way (except trying all combinations)?
-
- What about canonical co-ordinates? I always felt that this method
- gave you a better feel for what might be going on, but it is
- years since I did any econometrics.
-
- What are the drawbacks of this method, assuming you use it
- intelligently. (Obviously, the comments about the need for
- thought and intuition apply equally well whatever method
- you use).
-
-