groupedPMA-package: Penalized Multivariate Analysis

In ttriche/groupedPMA: Grouped Penalized Multivariate Analysis -- additional penalties for PMA

Description

The package is called groupedPMA, for "group Penalized Multivariate Analysis". It implements three methods: A penalized matrix decomposition, sparse principal components analysis, and sparse canonical correlations analysis. All are described in the paper "A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis", by D Witten, R Tibshirani, and T Hastie, and published in Biostatistics (2009). The paper is available at <http://www-stat.stanford.edu/~dwitten>.

The main functions are as follows: (1) PMD (2) CCA (3) SPC

The first function, PMD, performs a penalized matrix decomposition. CCA performs sparse canonical correlation analysis. SPC performs sparse principal components analysis.

There also are cross-validation functions for tuning parameter selection for each of the above methods: SPC.cv, PMD.cv, CCA.permute. PlotCGH results in nice plots for DNA copy number data.

GroupedPMA forks the originalgroupedPMA package to provide grouping penalties and automatic selection of shrinkage intensities based on StARS (Liu, 2010).

Details

Package:	groupedPMA
Type:	Package
Version:	1.0
Date:	2009-02-10
License:	GPL >= 2

Author(s)

Daniela M. Witten and Robert Tibshirani (edits by Tim Triche, Jr.)

References

Witten, Tibshirani and Hastie (2009) A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis. Biostatistics 10(3): 515-534. <http://www-stat.stanford.edu/~dwitten>

ttriche/groupedPMA documentation built on June 1, 2019, 2:51 a.m.