In this work we want to clarify, via a Monte Carlo experiment, if (and when) for an integer-valued time series it is really recommended to adopt the coherent forecasting methods from INAR models or if equivalently good predictions can be obtained from the simpler AR models. Results show that INAR models should be preferred.
Forecasting integer autoregressive processes of order 1: are simple AR competitive?
GEROLIMETTO, Margherita
2015-01-01
Abstract
In this work we want to clarify, via a Monte Carlo experiment, if (and when) for an integer-valued time series it is really recommended to adopt the coherent forecasting methods from INAR models or if equivalently good predictions can be obtained from the simpler AR models. Results show that INAR models should be preferred.File in questo prodotto:
File | Dimensione | Formato | |
---|---|---|---|
Bisaglia-Gerolimetto-EB2015.pdf
non disponibili
Tipologia:
Documento in Post-print
Licenza:
Accesso chiuso-personale
Dimensione
200.29 kB
Formato
Adobe PDF
|
200.29 kB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.