Welcome to Hyper-g Priors for GAM selection project!
This R-package implements objective Bayesian model selection with penalised splines, using a so-called hyper-g prior for the parameters in the generalised additive model.
Description:
The R-package hypergsplines implements objective Bayesian variable
selection in generalized additive models (Normal, Binomial and Poisson). As
objective parameters prior a (generalized) hyper-g prior is used. As objective
model prior so-called multiplicity correction priors are used. More details on
the methodology can be found in the
technical report (2011).
Download:
Currently the R-package hypergsplines is not available from CRAN,
because it is still under continuing development. The current versions are
available from R-Forge (this site). If you are running a recent R version, you
can obtain a binary snapshot of the development version as
install.packages("hypergsplines",repos="http://r-forge.r-project.org").
You can also manually download
the binary snapshot or
the source tarball.
You can have a look at the recent changes to the
package here.
The project summary page can be found here.
Documentation:
The R-package features a vignette which introduces the most important functions
in a logistic additive regression example. After installing the package, you can
open the vignette from within R with the command
vignette("examples", package="hypergsplines").
You can also have a look at the ISBA 2012 poster
presenting the methodology.
Developers:
Daniel Sabanes
Bove, Division of Biostatistics, Institute of Social and Preventive
Medicine, University of Zurich, Switzerland
