A fragmented-periodogram approach for clustering big data time series
Entity
UAM. Departamento de Análisis Económico: Economía CuantitativaPublisher
Springer NatureDate
2019-01-01Citation
10.1007/s11634-019-00365-8
Advances in Data Analysis and Classification x. June (2019): 1-30
ISSN
1862-5347 (print); 1862-5355 (online)DOI
10.1007/s11634-019-00365-8Funded by
Jorge Caiado and Nuno Crato have been supported through Project CEMAPREUID/ MULTI/00491/2019 financed by FCT/MCTES through national funds. This paper was finished when the Pilar Poncela returned to Universidad Autónoma de Madrid. Partial financial support from the Spanish Ministry of Economy and Competitiveness, Project ECO2015-70331-C2-1R and from Comunidad de Madrid, Project MadEco-CM S2015/HUM-3444 is acknowledged. Pilar Poncela, fellow of the UC3M-BS Institute of Financial Big Data (IFIBID), also thanks the Institute. Finally, we also thank Michela Nardo for providing the dataProject
Gobierno de España. ECO2015-70331-C2-1R; Comunidad de Madrid. S2015/HUM-3444/MADECOEditor's Version
https://doi.org/10.1007/s11634-019-00365-8Subjects
Big data; Fragmented periodogram; Smoothed periodogram; Spectral clustering; Time series clustering; EconomíaRights
© 2019, The Author(s)Abstract
We propose and study a new frequency-domain procedure for characterizing and comparing large sets of long time series. Instead of using all the information available from data, which would be computationally very expensive, we propose some regularization rules in order to select and summarize the most relevant information for clustering purposes. Essentially, we suggest to use a fragmented periodogram computed around the driving cyclical components of interest and to compare the various estimates. This procedure is computationally simple, but able to condense relevant information of the time series. A simulation exercise shows that the smoothed fragmented periodogram works in general better than the non-smoothed one and not worse than the complete periodogram for medium to large sample sizes. We illustrate this procedure in a study of the evolution of several stock markets indices. We further show the effect of recent financial crises over these indices behaviour.
Files in this item
Google Scholar:Caiado, Jorge
-
Crato, Nuno
-
Poncela Blanco, Pilar
This item appears in the following Collection(s)
Related items
Showing items related by title, author, creator and subject.