A characterizing property of Zenga (1984) inequality curve is exploited in order to develop an estimator for the extreme value index of a distribution with regularly varying tail. The approach proposed here has a nice graphical interpretation which provides a powerful method for the analysis of the tail of a distribution. The properties of the proposed estimation strategy are analysed theoretically and by means of simulations. The usefulness of the method will be tested also on real data sets.
Extreme value index estimation by means of an inequality curve / Taufer, Emanuele; Santi, Flavio; Inverardi, Pier Luigi Novi; Espa, Giuseppe; Dickson, Maria Michela. - In: MATHEMATICS. - ISSN 2227-7390. - ELETTRONICO. - 8:10(2020), pp. 1-17. [10.3390/math8101834]
Extreme value index estimation by means of an inequality curve
Taufer, EmanuelePrimo
;Santi, FlavioSecondo
;Inverardi, Pier Luigi Novi;Espa, GiuseppePenultimo
;Dickson, Maria MichelaUltimo
2020-01-01
Abstract
A characterizing property of Zenga (1984) inequality curve is exploited in order to develop an estimator for the extreme value index of a distribution with regularly varying tail. The approach proposed here has a nice graphical interpretation which provides a powerful method for the analysis of the tail of a distribution. The properties of the proposed estimation strategy are analysed theoretically and by means of simulations. The usefulness of the method will be tested also on real data sets.File | Dimensione | Formato | |
---|---|---|---|
2020 - Math - EV index estimation by means of an inequality curve.pdf
accesso aperto
Descrizione: Articolo completo
Tipologia:
Versione editoriale (Publisher’s layout)
Licenza:
Creative commons
Dimensione
1.13 MB
Formato
Adobe PDF
|
1.13 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione