[18850] in s-news-athena
Re: RE: [S] nls() fails in a loop
daemon@ATHENA.MIT.EDU (Prof Brian Ripley)
Wed Aug 4 09:55:32 1999
Message-Id: <199908041350.OAA13067@toucan.stats.ox.ac.uk>
Date: Wed, 4 Aug 1999 14:50:23 +0100 (BST)
From: Prof Brian Ripley <ripley@stats.ox.ac.uk>
Reply-To: Prof Brian Ripley <ripley@stats.ox.ac.uk>
To: internet:;internet:;@wubios.wustl.edu
Mime-Version: 1.0
Content-Type: TEXT/plain; charset=us-ascii
Content-Md5: fH53GFUKgOmNgJQl9o/hnQ==
> Date: Wed, 04 Aug 1999 08:32:58 -0400
> From: "Bruce McCullough" <BMCCULLO@fcc.gov>
>
> Bill Venables suggested:
>
> >1. Don't begin every cv fit with the original rough initial values. >Always
use the best available initial values you have, for >example, from the primary
fit using all the data.
>
> As another example, I have bootstrapped a few
> nonlinear models, and I have found that using
> the solution from the previous iteration can
> work well.
Can I suggest that neither approach is in general a valid one for a
cross-validation? If there are multiple local minima (there often are)
you are not cross-validating the original procedure (including choosing
the start) and so will underestimate (possibly severely) the
variability.
One approach Bill did not mention is to use self-starting nls models.
The version in version of nls in the nlme3.x library (and S-PLUS 2000)
is probably the easiest to use.
--
Brian D. Ripley, ripley@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272860 (secr)
Oxford OX1 3TG, UK Fax: +44 1865 272595
-----------------------------------------------------------------------
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
