mgcv
Updated 25/9/05. Changes are now listed in the changeLog file in the
source version of the package available on CRAN. For the history of older
changes see
update list. An overview for
people familiar with gam() in S-PLUS is provided.
This page contains an
R
package for performing
multiple smoothing parameter estimation for generalized ridge regression
problems by GCV/UBRE and for working with thin-plate regression splines and low rank tensor product smooths. In particular it provides a version of gam() for R, with (optional) integrated smoothness estimation, and a gamm() function for generalized additive mixed modelling.
The GCV method generalizes an approach first suggested by Gu & Wahba
(1991),
and is reported in Wood (2000) JRSSB 62:159-174 and Wood (2004) JASA 99:637-686. Thin-plate regression splinesare described in
Wood (2003) JRSSB 65:95-114 and the tensor product smooths are described in
Wood (2004)
tech report of the university of Glasgow (Biometrics, 2006). The
package also contains C code that can be used on its own.
(Examples for routine pcls also include monotonic regression.)
If you have R installed you probably
have mgcv installed already: just type library(mgcv) at the R
command prompt to find out (and check the version number).
The code and all related documentation are provided under the
GNU General Public License (2)
- To be sure of having the latest release version, it's best to download it from:
CRAN.
- Here is a .zip file containing
pdf slide presentations from a short course given at useR 2006, along with
LaTex source and original .eps files.
- Please let me know if you have any problems installing or using
mgcv: it's intended to work, so I'd like to know if it doesn't! My
email is: s.wood _at_ bath.ac.uk
Henric Nilsson has kindly donated code that substantially improved
summary.gam
and related functions and plot.gam.
Thanks to the following (incomplete list of) people for bug reports suggestions and help.
Nicole Augustin; Mark Bravington; Louise Burt; Liz Clarke; Mark Clements;
Peter Dalgaard;
Anthony Davison;Sharon Hedley;
Kurt Hornik;Pierre Joyet; Andy Liaw; Thomas Maiwald;
Henric Nilsson; Jari Oksanen; Charles Paxton; Greg Ridgeway; Brian Ripley;
Evi Samoli; John Szumiloski;
Alain Le Tertre; Luke Tierney; Brian Williams; Jim Young.
Finally, I am particularly grateful to David Borchers and Chong Gu (anonymously!) for first suggesting
making these methods available in S and Mike Lonergan for a good deal of helpful discussion and many
useful suggestions
about numerous aspects of the package (including the idea for and earlier code for vis.gam, and the
earlier versions of the negative binomial code.)
Simon N. Wood simon@stats.gla.ac.uk
Home |
Back