Circulant singular spectrum analysis: A new automated procedure for signal extraction
EntityUAM. Departamento de Análisis Económico: Economía Cuantitativa
10.1016/j.sigpro.2020.107824Signal Processing 179 (2021): 107824
ISSN0165-1684 (print); 1872-7557 (online)
Funded byFinancial support from the Spanish government, contract grants MINECO/FEDER ECO2015-70331-C2-1-R, ECO2015-66593-P, ECO2016-76818-C3-3-P, PID2019-107161GB-C32 and PID2019-108079GB-C22 is acknowledged
ProjectGobierno de España. ECO2015-70331-C2-1-R; Gobierno de España. PID2019-108079GB-C22
SubjectsAM-FM Signals; Circulant matrices; Principal components; Signal extraction; Singular spectrum analysis; Singular value decomposition; Economía
Rights© 2020 The Authors
Esta obra está bajo una licencia de Creative Commons Reconocimiento-NoComercial-SinObraDerivada 4.0 Internacional.
Sometimes, it is of interest to single out the fluctuations associated to a given frequency. We propose a new variant of SSA, Circulant SSA (CiSSA), that allows to extract the signal associated to any frequency specified beforehand. This is a novelty when compared with other SSA procedures that need to identify ex-post the frequencies associated to the extracted signals. We prove that CiSSA is asymptotically equivalent to these alternative procedures although with the advantage of avoiding the need of the subsequent frequency identification. We check its good performance and compare it to alternative SSA methods through several simulations for linear and nonlinear time series. We also prove its validity in the nonstationary case. We apply CiSSA in two different fields to show how it works with real data and find that it behaves successfully in both applications. Finally, we compare the performance of CiSSA with other state of the art techniques used for nonlinear and nonstationary signals with amplitude and frequency varying in time
This item appears in the following Collection(s)
Showing items related by title, author, creator and subject.