|
The
S+BEST library contains functions similar to (because they are derived
from) software by
Charles Kooperberg including software for logspline density estimation,
hazard regression, and
adaptive generalized linear models. The "regression" family in the
adaptive generalized linear models
function fits models si milar in character to MARS. This software
can be access on Linux, windows, and
solaris environments for S-PLUS 6 by down loading the S+BEST library from
Doug
Clarkson
-----Original Message-----
From:
mzhang@cshl.org [mailto:mzhang@cshl.org]
Sent: Thu 1/23/2003 7:15
AM
To: s-news@lists.biostat.wustl.edu
Cc:
mzhang@cshl.edu
Subject: [S] Summary on MARS
I want to Thank Any Liaw, Tamara Shatar, Rob Tibshirani,
Patricia Beziat, Eric Wong,
Sean Keenan, Falk Huettmann, Kunio Takezawa,
Nick Ellis, Bjxrn-Helge Mevik,
and others.
Here are some sample of
answers:
The library mda by Trevor Hastie and Robert Tibshirani
contains an
implementation of mars.
For Splus go to http://www.stats.ox.ac.uk/pub/MASS4/Winlibs/.
For
R simply load the package from CRAN.
-Nick Ellis
Try the website
http://www-stat.stanford.edu/~jhf/MART.html
<http://www-stat.stanford.edu/~jhf/MART.html>
Cheers,
-Sean
Keenan
T
revor Hastie has implemented mars in the mda library, which
can be found
either on STATLIB or Hastie's web site. The same code
has been ported to
the `mda' package for R (available on
CRAN).
Charles Kooperberg also has an R package `polspline' that fits
similar
models to MARS, but has more features (e.g., estimates density,
hazard rate,
classification, etc.)
-Andy Liaw
Venables and
Ripley discuss this in Modern Applied Statistics with S (4th
edition -
page 235). Apparently it is in library section mda or
another
restricted version is in library section polymers (Kooperberg and
O'Connor).
-Tamara Shatar.
-michael
--
Michael Q.
Zhang
Professor
Phone: (516)
367-8393
Watson School of Biological Sciences
Fax: (516) 367-8461 or
6862 (Winfax) Cold Spring Harbor
Laboratory
E-mail:
mzhang@cshl.edu
1 Bungtown Road,
Secretary: Ms. Carol Marcincuk
367-8387 P.O.Box 100,
WebURL: http://www.cshl.edu/mzhanglab
Cold Spring Harbor, NY 11724
USA
--------------------------------------------------------------------
This
message was distributed by s-news@lists.biostat.wustl.edu.
To
unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu
with
the BODY of the message: unsubscribe
s-news
|