In this work we show how to use the R package yuima available on CRAN for the estimation of a Continuous Autoregressive Moving Average (CARMA) model on the real data. When dealing with the CARMA model, one of the advantages of the yuima package is the possibility of recovering the increments of the underlying noise and choosing the appropriate Lévy model. The estimation of the parameters for the underlying L´evy process makes yuima package appealing for modeling financial time series. Indeed, identifying the appropriate noise for a CARMA model allows to capture asymmetry and heavy tails observed in the real data.
Estimation of Lévy CARMA models in the yuima package: application on the financial time series / S.M. Iacus, L. Mercuri - In: Proceedings of COMPSTAT 2014 21st International Conference on Computational Statistics hosting the 5th IASC World Conference / [a cura di] M. Gilli, G. Gonzalez-Rodriguez, A. Nieto-Reyes. - [s.l] : ISI/IASC, 2014 Aug. - ISBN 978-2-8399-1347-8. - pp. 451-458 (( convegno Proceedings of COMPSTAT 2014 21st International Conference on Computational Statistics hosting the 5th IASC World Conference tenutosi a Geneva nel 2014.
Estimation of Lévy CARMA models in the yuima package: application on the financial time series
S.M. Iacus;L. Mercuri
2014
Abstract
In this work we show how to use the R package yuima available on CRAN for the estimation of a Continuous Autoregressive Moving Average (CARMA) model on the real data. When dealing with the CARMA model, one of the advantages of the yuima package is the possibility of recovering the increments of the underlying noise and choosing the appropriate Lévy model. The estimation of the parameters for the underlying L´evy process makes yuima package appealing for modeling financial time series. Indeed, identifying the appropriate noise for a CARMA model allows to capture asymmetry and heavy tails observed in the real data.File | Dimensione | Formato | |
---|---|---|---|
ErcimIacus.pdf
solo utenti autorizzati
Tipologia:
Publisher's version/PDF
Dimensione
1.99 MB
Formato
Adobe PDF
|
1.99 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.