[R-pkgs] New package: BiplotGUI
La Grange, AM
amlg at sun.ac.za
Wed Aug 13 12:04:20 CEST 2008
Dear R users,
I am pleased to announce the release of the new BiplotGUI package on
CRAN.
Biplots are graphs in which the samples and all the variables of a data
matrix are represented simultaneously. They can be very useful for
exploring multivariate data.
The BiplotGUI package allows users to construct and interact with
biplots as they are described in "Biplots" (Gower and Hand, 1996). In
such biplots, the samples are represented as points, while the variables
are represented as calibrated biplot axes. The representations are
similar to ordinary scatter plots, but with all the variables
represented at one time.
Version 0.0-2 of the package has the following features:
* A graphical user interface (GUI), making it easy to use
* Support for different types of biplots (PCA, covariance/correlation,
CVA, regression, Procrustes, circular non-linear), both interpolative
and predictive
* Support for different scaling techniques (PCO, metric MDS, non-metric
MDS, semi-metric MDS)
* Various data transformations and distance metrics
* Additional descriptors (convex hulls, alpha-bags, point densities,
classification regions)
* Intermediate quantities for export back to R
* Diagnostics (graphs of convergence; point, group and axis
predictivities; Shepard diagrams)
* Interactivity (variable value prediction, zooming, point and axis
drag-and-drop, 3D biplots)
* Highly customisable graphics
* Import of data directly from Excel 1997-2003 files
The package can be downloaded from CRAN:
<http://cran.r-project.org/web/packages/BiplotGUI/index.html> or a
favourite mirror.
The project homepage is at <http://biplotgui.r-forge.r-project.org>. It
includes screenshots and example biplots. A manual is included within
the package itself, but is also available from the website.
At present, the package runs under Windows only.
I would appreciate any feedback so that the package can be further
improved.
Thanks,
Anthony la Grange
Department of Statistics and Actuarial Science
Stellenbosch University
South Africa
More information about the R-packages
mailing list