The "whitening" package implements the whitening methods (ZCA, PCA, Cholesky, ZCA-cor, and PCA-cor) discussed in Kessy, Lewin, and Strimmer (2018) as well as the whitening approach to Canonical Correlation Analysis (CCA) allowing negative canonical correlations described in Jendoubi and Strimmer (2018).
Current Version: 1.0.0
Authors: Korbinian Strimmer, Takoua Jendoubi, Agnan Kessy, and Alex Lewin.
Documentation and Installation:
Quick install: enter at the R console:
install.packages("whitening")
- Manual (pdf file) and release history.
- Download of whitening version 1.0.0 source package.
- Archive of previous versions of whitening.
- Licensed under the GNU GPL version 3 (or any later version).
Publications:
- A. Kessy, A. Lewin, and K. Strimmer. 2018. Optimal whitening and decorrelation. The American Statistician, in press. (arXiv:1512.00809)
- T. Jendoubi and K. Strimmer. 2018. Probabilistic canonical correlation analysis: a whitening approach. (arXiv:1802.03490)
R Code for Kessy, Lewin, and Strimmer (2018):
- The five discussed natural whitening procedures are implemented in the two functions
whiteningMatrix
andwhitening
. - You can also download R code for reproducing the results in Table 2 presented in the paper.
R Code for Jendoubi and Strimmer (2018):
- The whitening approach to canonical correlation analysis is implemented in the functions
cca
(empirial estimator) andscca
(shrinkage estimator). - R code to reproduce the nutrimouse data analysis.
- R code to reproduce the NCI60 data analysis.