We present a procedure for clustering time series according to their tail dependence behaviour as measured via a suitable copula-based tail coefficient, estimated in a non-parametric way. Simulation results about the proposed methodology together with an application to financial data are presented showing the usefulness of the proposed approach.

Clustering of time series via non-parametric tail dependence estimation

PAPPADA' , ROBERTA;TORELLI, Nicola
2015-01-01

Abstract

We present a procedure for clustering time series according to their tail dependence behaviour as measured via a suitable copula-based tail coefficient, estimated in a non-parametric way. Simulation results about the proposed methodology together with an application to financial data are presented showing the usefulness of the proposed approach.
13-giu-2014
Pubblicato
https://link.springer.com/article/10.1007%2Fs00362-014-0605-7
File in questo prodotto:
File Dimensione Formato  
art%3A10.1007%2Fs00362-014-0605-7.pdf

Accesso chiuso

Tipologia: Documento in Versione Editoriale
Licenza: Copyright Editore
Dimensione 619.93 kB
Formato Adobe PDF
619.93 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
2787726_art_10.1007_s00362-014-0605-7-PostPrint.pdf

accesso aperto

Descrizione: PostPrint VQR3
Tipologia: Bozza finale post-referaggio (post-print)
Licenza: Digital Rights Management non definito
Dimensione 930.29 kB
Formato Adobe PDF
930.29 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11368/2787726
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 20
  • ???jsp.display-item.citation.isi??? 19
social impact