Disentangling dynamic and stochastic modes in multivariate time series

A signal decomposition is presented that disentangles the deterministic and stochastic components of a multivariate time series. The dynamical component analysis (DyCA) algorithm is based on the assumption that an unknown set of ordinary differential equations (ODEs) describes the dynamics of the de...

Full description

Saved in:
Bibliographic Details
Main Authors: Christian Uhl, Annika Stiehl, Nicolas Weeger, Markus Schlarb, Knut Hüper
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-10-01
Series:Frontiers in Applied Mathematics and Statistics
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fams.2024.1456635/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850214980782129152
author Christian Uhl
Annika Stiehl
Nicolas Weeger
Markus Schlarb
Markus Schlarb
Knut Hüper
author_facet Christian Uhl
Annika Stiehl
Nicolas Weeger
Markus Schlarb
Markus Schlarb
Knut Hüper
author_sort Christian Uhl
collection DOAJ
description A signal decomposition is presented that disentangles the deterministic and stochastic components of a multivariate time series. The dynamical component analysis (DyCA) algorithm is based on the assumption that an unknown set of ordinary differential equations (ODEs) describes the dynamics of the deterministic part of the signal. The algorithm is thoroughly derived and accompanied by a link to the GitHub repository containing the algorithm. The method was applied to both simulated and real-world data sets and compared to the results of principal component analysis (PCA), independent component analysis (ICA), and dynamic mode decomposition (DMD). The results demonstrate that DyCA is capable of separating the deterministic and stochastic components of the signal. Furthermore, the algorithm is able to estimate the number of linear and non-linear differential equations and to extract the corresponding amplitudes. The results demonstrate that DyCA is an effective tool for signal decomposition and dimension reduction of multivariate time series. In this regard, DyCA outperforms PCA and ICA and is on par or slightly superior to the DMD algorithm in terms of performance.
format Article
id doaj-art-534ae9ce2fb946a099fd2a8a0db4b29a
institution OA Journals
issn 2297-4687
language English
publishDate 2024-10-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Applied Mathematics and Statistics
spelling doaj-art-534ae9ce2fb946a099fd2a8a0db4b29a2025-08-20T02:08:45ZengFrontiers Media S.A.Frontiers in Applied Mathematics and Statistics2297-46872024-10-011010.3389/fams.2024.14566351456635Disentangling dynamic and stochastic modes in multivariate time seriesChristian Uhl0Annika Stiehl1Nicolas Weeger2Markus Schlarb3Markus Schlarb4Knut Hüper5Center for Signal Analysis of Complex Systems, Ansbach University of Applied Sciences, Ansbach, GermanyCenter for Signal Analysis of Complex Systems, Ansbach University of Applied Sciences, Ansbach, GermanyCenter for Signal Analysis of Complex Systems, Ansbach University of Applied Sciences, Ansbach, GermanyCenter for Signal Analysis of Complex Systems, Ansbach University of Applied Sciences, Ansbach, GermanyInstitute of Mathematics, Julius-Maximilians-Universität, Würzburg, GermanyInstitute of Mathematics, Julius-Maximilians-Universität, Würzburg, GermanyA signal decomposition is presented that disentangles the deterministic and stochastic components of a multivariate time series. The dynamical component analysis (DyCA) algorithm is based on the assumption that an unknown set of ordinary differential equations (ODEs) describes the dynamics of the deterministic part of the signal. The algorithm is thoroughly derived and accompanied by a link to the GitHub repository containing the algorithm. The method was applied to both simulated and real-world data sets and compared to the results of principal component analysis (PCA), independent component analysis (ICA), and dynamic mode decomposition (DMD). The results demonstrate that DyCA is capable of separating the deterministic and stochastic components of the signal. Furthermore, the algorithm is able to estimate the number of linear and non-linear differential equations and to extract the corresponding amplitudes. The results demonstrate that DyCA is an effective tool for signal decomposition and dimension reduction of multivariate time series. In this regard, DyCA outperforms PCA and ICA and is on par or slightly superior to the DMD algorithm in terms of performance.https://www.frontiersin.org/articles/10.3389/fams.2024.1456635/fulldynamical component analysis (DyCA)dynamic mode decomposition (DMD)dimension reductiondynamical systemsblind source separationdifferential equations
spellingShingle Christian Uhl
Annika Stiehl
Nicolas Weeger
Markus Schlarb
Markus Schlarb
Knut Hüper
Disentangling dynamic and stochastic modes in multivariate time series
Frontiers in Applied Mathematics and Statistics
dynamical component analysis (DyCA)
dynamic mode decomposition (DMD)
dimension reduction
dynamical systems
blind source separation
differential equations
title Disentangling dynamic and stochastic modes in multivariate time series
title_full Disentangling dynamic and stochastic modes in multivariate time series
title_fullStr Disentangling dynamic and stochastic modes in multivariate time series
title_full_unstemmed Disentangling dynamic and stochastic modes in multivariate time series
title_short Disentangling dynamic and stochastic modes in multivariate time series
title_sort disentangling dynamic and stochastic modes in multivariate time series
topic dynamical component analysis (DyCA)
dynamic mode decomposition (DMD)
dimension reduction
dynamical systems
blind source separation
differential equations
url https://www.frontiersin.org/articles/10.3389/fams.2024.1456635/full
work_keys_str_mv AT christianuhl disentanglingdynamicandstochasticmodesinmultivariatetimeseries
AT annikastiehl disentanglingdynamicandstochasticmodesinmultivariatetimeseries
AT nicolasweeger disentanglingdynamicandstochasticmodesinmultivariatetimeseries
AT markusschlarb disentanglingdynamicandstochasticmodesinmultivariatetimeseries
AT markusschlarb disentanglingdynamicandstochasticmodesinmultivariatetimeseries
AT knuthuper disentanglingdynamicandstochasticmodesinmultivariatetimeseries