Abstract
The problem of time series clustering has attracted growing research interest in the last decade. The most popular clustering methods assume that the time series are only linearly dependent but this assumption usually fails in practice. To overcome this limitation, in this paper, we study clustering methods applicable to time series with a general dependent (possibly nonlinear) structure. We propose a dissimilarity measure based on the auto distance correlation function which is able to detect both linear and nonlinear dependence structures. Once the pairwise dissimilarity matrix for time series has been obtained, a standard clustering algorithm, such as hierarchical clustering algorithm, can be used. Numerical studies based on Monte Carlo experiments show that our method performs reasonably well.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Fokianos, F., Pitsillou, M.: Consistent testing for pairwise dependence in time series. Technometrics 159, 262–270 (2017)
Fokianos, F., Pitsillou, M.: Testing independence for multivariate time series via the auto-distance correlation matrix. Biometrika 105, 337–352 (2018)
Zhou, Z.: Measuring nonlinear dependence in time-series, a distance correlation approach. J. Time Ser. Anal. 33, 438–457 (2012)
Pitsillou, M., Fokianos, F.: dCovTS: distance covariance/correlation for time series. R J. 8, 324–340 (2016)
Alonso, A.M., Pena, D.: Clustering time series by linear dependency. Stat. Comput. 19, 655–676 (2019)
Lange, T., Roth, V., Braun, M.L., Buhmann, J.M.: Stability-based validation of clustering solutions. Neural Comput. 16, 1299–1323 (2004)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
La Rocca, M., Vitale, L. (2021). Clustering Time Series by Nonlinear Dependence. In: Corazza, M., Gilli, M., Perna, C., Pizzi, C., Sibillo, M. (eds) Mathematical and Statistical Methods for Actuarial Sciences and Finance. Springer, Cham. https://doi.org/10.1007/978-3-030-78965-7_43
Download citation
DOI: https://doi.org/10.1007/978-3-030-78965-7_43
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-78964-0
Online ISBN: 978-3-030-78965-7
eBook Packages: Economics and FinanceEconomics and Finance (R0)