ForeCA: ForeCA - Forecastable Component Analysis

ForeCA is a novel dimension reduction (DR) technique for temporally dependent signals. Contrary to other popular DR methods, such as PCA or ICA, ForeCA explicitly searches for the most ”forecastable” signal. The measure of forecastability Omega(x_t) is based on negative Shannon entropy of the spectral density of the transformed signal. This R package provides the main algorithms and some auxiliary function (summary, plotting, etc).

Version: 0.0.9
Depends: R (≥ 2.12.0), sapa, mgcv, astsa, class
Imports: mgcv, sapa, R.utils
Published: 2012-12-12
Author: Georg M. Goerg
Maintainer: Georg M. Goerg <gmg at stat.cmu.edu>
License: GPL (≥ 2)
URL: http://www.stat.cmu.edu/~gmg http://arxiv.org/abs/1205.4591
NeedsCompilation: no
In views: TimeSeries
CRAN checks: ForeCA results

Downloads:

Package source: ForeCA_0.0.9.tar.gz
MacOS X binary: ForeCA_0.0.9.tgz
Windows binary: ForeCA_0.0.9.zip
Reference manual: ForeCA.pdf
News/ChangeLog:NEWS
Old sources: ForeCA archive