POCRE: Penalized Orthogonal-Components Regression

Penalized orthogonal-components regression (POCRE) is a supervised dimension reduction method for high-dimensional data. It sequentially constructs orthogonal components (with selected features) which are maximally correlated to the response residuals. POCRE can also construct common components for multiple responses and thus build up latent-variable models.

Version: 0.5.0
Imports: stats, utils, ggplot2 (≥ 2.2.0), pracma, EbayesThresh
Published: 2018-07-20
Author: Dabao Zhang, Zhongli Jiang, Zeyu Zhang
Maintainer: Dabao Zhang <zhangdb at purdue.edu>
License: GPL-2
NeedsCompilation: no
CRAN checks: POCRE results


Reference manual: POCRE.pdf
Package source: POCRE_0.5.0.tar.gz
Windows binaries: r-devel: POCRE_0.5.0.zip, r-release: POCRE_0.5.0.zip, r-oldrel: POCRE_0.5.0.zip
macOS binaries: r-release: POCRE_0.5.0.tgz, r-oldrel: POCRE_0.5.0.tgz


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