odpc: One-Sided Dynamic Principal Components

Functions to compute the one-sided dynamic principal components ('odpc') introduced in Smucler, Peña and Yohai (2017) <arXiv:1708.04705>. 'odpc' is a novel dimension reduction technique for multivariate time series, that is useful for forecasting. These dynamic principal components are defined as the linear combinations of the present and past values of the series that minimize the reconstruction mean squared error.

Version: 1.0.0
Depends: R (≥ 3.3.1)
Imports: methods, Rcpp (≥ 0.12.7), gdpc, forecast
LinkingTo: Rcpp, RcppArmadillo (≥ 0.7.500.0.0)
Published: 2017-08-18
Author: Daniel Peña, Ezequiel Smucler, Victor Yohai
Maintainer: Ezequiel Smucler <ezequiels.90 at gmail.com>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
In views: TimeSeries
CRAN checks: odpc results

Downloads:

Reference manual: odpc.pdf
Package source: odpc_1.0.0.tar.gz
Windows binaries: r-devel: odpc_1.0.0.zip, r-release: odpc_1.0.0.zip, r-oldrel: odpc_1.0.0.zip
OS X El Capitan binaries: r-release: odpc_1.0.0.tgz
OS X Mavericks binaries: r-oldrel: odpc_1.0.0.tgz

Linking:

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