Graphical
modelling packages (for R)
- gRain
-
gRapical Independence Networks in R:
A package for probability propagation in graphical independence
networks.
- mimR -
graphical
modelling in R: An R interface
to the MIM program for graphical modelling. Note: MIM is available on
Windows only.
- gRim -
gRaphical Interaction
Models in R: An experimental
package (not on CRAN) for log-linear models and graphical Gaussian
models.
- gRbase:
The base package for the other graphical modelling packages. Provides
no essential modelling functionality from a users perspective but the
package is used by the
other packages. The package does contain several graph algoritms. Will
be installed automatically when installing one of
the other packages.
- gRc: Inference
in Graphical Gaussian Models with Edge and Vertex
Symmetries. Estimation, model selection and other aspects of
statistical inference in Graphical Gaussian models with edge and vertex
symmetries (Graphical Gaussian models with colours).
Optional
(but
highly recommended): Rgraphviz
These
R-packages use the R-package Rgraphviz as well as the
stand-alone program graphviz for displaying graphs. For
installation instructions please see http://rhugin.r-forge.r-project.org/.
The Rgraphviz
package must be loaded
manually using library(Rgraphviz)
Other
R-packages:
- doBy: Facilities
for groupwise computations of summary statistics and other facilities
for working with grouped data (similar to what can be achieved by proc
means or proc summary of the sas system).
- geepack:
Generalized Estimating Equation Package. Generalized estimating
equations solver for parameters in mean, scale, and correlation
structures, through mean link, scale link, and correlation link. Can
also handle clustered categorical responses.
- Ryacas:
R interface to the yacas computer algebra system.See http://ryacas.googlecode.com