[R] will gam() be available soon?
Prof Brian Ripley
ripley at stats.ox.ac.uk
Fri Dec 29 17:20:14 CET 2000
On Fri, 29 Dec 2000, Sung-il Cho wrote:
> I wonder if generalized additive model (gam) capability
> will be available in R sooner or later?
> Anybody know about any ongoing project or plans on this?
Depends exactly what you want. There is a gam function in package
mgcv, but it is not an exact clone (no lo() for example). Package gss
can fit spline-based gams too. And if you can accept regression splines
you can use glm. For Gaussian gams you can use bruto from package mda.
It you really want gam() as in S, please use S-PLUS (you must have access
to it given your address). There are no plans to implement all its quirks
You will find collections of plans on the developer.r-project.org site,
Something better than gam for additive models. Perhaps
along the lines of BRUTO (in package mda).
One issue is selection of smoothness, which S's gam does very
slowly and unreliably.
You will also find in the list archive powerful arguments against any
use of additive models from a recent discussion on this.
(Kurt is planning to add something to the FAQ about this, but that depends
in part on our understanding the precise limitations of what we currently
have, and the packages keep on being upgraded. I still find
mgcv's gam fails most times I try it, but your mileage may differ.)
Brian D. Ripley, ripley at 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
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
More information about the R-help