coroICA: Confounding Robust Independent Component Analysis for Noisy and Grouped Data

Contains an implementation of a confounding robust independent component analysis (ICA) for noisy and grouped data. The main function coroICA() performs a blind source separation, by maximizing an independence across sources and allows to adjust for varying confounding based on user-specified groups. Additionally, the package contains the function uwedge() which can be used to approximately jointly diagonalize a list of matrices. For more details see the project website <https://sweichwald.de/coroICA/>.

Version: 1.0.1
Depends: R (≥ 3.2.3)
Imports: stats, MASS
Published: 2018-12-30
Author: Niklas Pfister and Sebastian Weichwald
Maintainer: Niklas Pfister <pfister at stat.math.ethz.ch>
BugReports: https://github.com/sweichwald/coroICA-R/issues
License: AGPL-3
URL: https://github.com/sweichwald/coroICA-R
NeedsCompilation: no
CRAN checks: coroICA results

Downloads:

Reference manual: coroICA.pdf
Package source: coroICA_1.0.1.tar.gz
Windows binaries: r-devel: coroICA_1.0.1.zip, r-release: coroICA_1.0.1.zip, r-oldrel: coroICA_1.0.1.zip
OS X binaries: r-release: coroICA_1.0.1.tgz, r-oldrel: not available

Linking:

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