ForeCA: Forecastable Component Analysis

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 takes time dependency explicitly into account and searches for the most ”forecastable” signal. The measure of forecastability is based on the Shannon entropy of the spectral density of the transformed signal. This R package provides the main algorithms and auxiliary function (summary, plotting, etc.) to apply ForeCA to multivariate time series data.

Version: 0.2.0
Depends: R (≥ 3.0.0), ifultools (≥ 2.0.0)
Imports: MASS, sapa
Suggests: astsa, mgcv, nlme (≥ 3.1-64), testthat (≥ 0.9.0), rSFA
Published: 2015-02-20
Author: Georg M. Goerg
Maintainer: Georg M. Goerg <im at gmge.org>
License: GPL-2
URL: http://www.gmge.org
NeedsCompilation: no
Citation: ForeCA citation info
Materials: NEWS
In views: TimeSeries
CRAN checks: ForeCA results

Downloads:

Reference manual: ForeCA.pdf
Package source: ForeCA_0.2.0.tar.gz
Windows binaries: r-devel: ForeCA_0.2.0.zip, r-release: ForeCA_0.2.0.zip, r-oldrel: ForeCA_0.2.0.zip
OS X Snow Leopard binaries: r-release: ForeCA_0.2.0.tgz, r-oldrel: ForeCA_0.2.0.tgz
OS X Mavericks binaries: r-release: ForeCA_0.2.0.tgz
Old sources: ForeCA archive