A1 Refereed original research article in a scientific journal

New independent component analysis tools for time series



Authors: Markus Matilainen, Klaus Nordhausen, Hannu Oja

Publisher: Elsevier

Publication year: 2015

Journal: Statistics and Probability Letters

Volume: 105

First page : 80

Last page: 87

Number of pages: 8

ISSN: 0167-7152

DOI: https://doi.org/10.1016/j.spl.2015.04.033

Web address : http://www.sciencedirect.com/science/article/pii/S0167715215001868


Abstract

Independent component analysis is a popular approach in search of latent variables and structures in high-dimensional data. We propose extensions of classical FOBI and JADE estimates for multivariate time series, with a special focus on time series with stochastic volatility.




Last updated on 2024-26-11 at 11:46