PCDimension: Finding the Number of Significant Principal Components

Implements methods to automate the Auer-Gervini graphical Bayesian approach for determining the number of significant principal components. Automation uses clustering, change points, or simple statistical models to distinguish "long" from "short" steps in a graph showing the posterior number of components as a function of a prior parameter. See <doi:10.1101/237883>.

Version: 1.1.8
Depends: R (≥ 3.1), ClassDiscovery
Imports: methods, stats, graphics, oompaBase, kernlab, changepoint, cpm
Suggests: MASS, nFactors
Published: 2018-01-09
Author: Kevin R. Coombes, Min Wang
Maintainer: Kevin R. Coombes <krc at silicovore.com>
License: Apache License (== 2.0)
URL: http://oompa.r-forge.r-project.org/
NeedsCompilation: no
Materials: NEWS
CRAN checks: PCDimension results

Downloads:

Reference manual: PCDimension.pdf
Vignettes: PCDimension
Package source: PCDimension_1.1.8.tar.gz
Windows binaries: r-devel: PCDimension_1.1.8.zip, r-release: PCDimension_1.1.8.zip, r-oldrel: PCDimension_1.1.8.zip
OS X El Capitan binaries: r-release: PCDimension_1.1.8.tgz
OS X Mavericks binaries: r-oldrel: PCDimension_1.1.7.tgz
Old sources: PCDimension archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=PCDimension to link to this page.