The current state of the art in selecting ARMA time series models requires competence and experience on the part of the practitioner, and sometimes the results are not very satisfactory. In this paper, we propose a new automatic approach to the model selection problem, based upon evolutionary computation. We build a genetic algorithm which evolves the representation of a predictive model, choosing both the orders and the predictors of the model. In simulation studies, the procedure succeeded in identifying the data generating process in the great majority of cases studied.

Building ARMA models with genetic algorithms / Minerva, Tommaso; Poli, I.. - 2037:(2001), pp. 335-342. ((Intervento presentato al convegno EvoWorkshops 2000 tenutosi a COMO, ITALY nel APR 18-20, 2001 [10.1007/3-540-45365-2_35].

Building ARMA models with genetic algorithms

MINERVA, Tommaso;
2001-01-01

Abstract

The current state of the art in selecting ARMA time series models requires competence and experience on the part of the practitioner, and sometimes the results are not very satisfactory. In this paper, we propose a new automatic approach to the model selection problem, based upon evolutionary computation. We build a genetic algorithm which evolves the representation of a predictive model, choosing both the orders and the predictors of the model. In simulation studies, the procedure succeeded in identifying the data generating process in the great majority of cases studied.
EvoWorkshops 2000
COMO, ITALY
APR 18-20, 2001
2037
335
342
Minerva, Tommaso; Poli, I.
Building ARMA models with genetic algorithms / Minerva, Tommaso; Poli, I.. - 2037:(2001), pp. 335-342. ((Intervento presentato al convegno EvoWorkshops 2000 tenutosi a COMO, ITALY nel APR 18-20, 2001 [10.1007/3-540-45365-2_35].
File in questo prodotto:
File Dimensione Formato  
3-540-45365-2_35.pdf

Accesso riservato

Tipologia: Versione pubblicata dall'editore
Dimensione 107.41 kB
Formato Adobe PDF
107.41 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/6070
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 23
  • ???jsp.display-item.citation.isi??? 20
social impact