odpc (2.0.1)

One-Sided Dynamic Principal Components.


Functions to compute the one-sided dynamic principal components ('odpc') introduced in Smucler, Pea and Yohai (2018) . '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.

Maintainer: Ezequiel Smucler
Author(s): Daniel Pea <daniel.pena@uc3m.es>, Ezequiel Smucler <ezequiels.90@gmail.com>, Victor Yohai <vyohai@dm.uba.ar>

License: GPL (>= 2)

Uses: doParallel, foreach, forecast, Rcpp, testthat

Released 3 days ago.