This work exploits the potential of bootstrap methods in testing for serial dependence in the class of observed driven Integer-AutoRegressive (INAR) models with Poisson innovations. The main contribution is to develop a novel restricted bootstrap algorithm to improve the performance of score-based test statistic, especially in case of moderately small series.
Bootstrap-based score test for INAR effect / Ievoli, Riccardo; Palazzo, Lucio. - (2021), pp. 1581-1586. (Intervento presentato al convegno 50th edition of the Scientific Meeting of the Italian Statistical Society tenutosi a Pisa nel 21-25 giugno 2021).
Bootstrap-based score test for INAR effect
Palazzo, Lucio
2021
Abstract
This work exploits the potential of bootstrap methods in testing for serial dependence in the class of observed driven Integer-AutoRegressive (INAR) models with Poisson innovations. The main contribution is to develop a novel restricted bootstrap algorithm to improve the performance of score-based test statistic, especially in case of moderately small series.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
Ievoli_Palazzo_SIS21.pdf
non disponibili
Tipologia:
Versione Editoriale (PDF)
Licenza:
Accesso privato/ristretto
Dimensione
99.22 kB
Formato
Adobe PDF
|
99.22 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
