Title: A fragmented-periodogram approach for clustering big data time series
Authors: CAIADO JORGECRATO NUNOPONCELA BLANCO MARIA DEL PILAR
Citation: ADVANCES IN DATA ANALYSIS AND CLASSIFICATION vol. 14 no. 1 p. 117-146
Publisher: SPRINGER HEIDELBERG
Publication Year: 2020
JRC N°: JRC112710
ISSN: 1862-5347 (online)
URI: https://link.springer.com/article/10.1007%2Fs11634-019-00365-8
https://publications.jrc.ec.europa.eu/repository/handle/JRC112710
DOI: 10.1007/s11634-019-00365-8
Type: Articles in periodicals and books
Abstract: We propose and study a new frequency-domain procedure for characterizing and comparing large sets of long financial 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 propose to use a fragmented periodogram computed around the driving seasonal components of interest and compare the various estimates. This procedure is computationally simple, but able to condense relevant second-order information of the volatility of the time series. A simulation exercise shows that the smoothed fragmented periodogram works in general better than the non-smoothed one and not significantly worse than the complete periodogram. We apply this procedure to study the evolution of several stock markets indices, extracting information on the European financial integration. We further show the effect of the recent financial crisis over these indices behaviour.
JRC Directorate:Growth and Innovation

Files in This Item:
There are no files associated with this item.


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.