Performance of ARIMA models in time series / by Kim Chiu ; John Higginson ; Guy Huot. : CS11-614/85-22E-PDF

"The purpose of this paper is to study a set of eight criteria which when applied to the Box-Jenkins method permit an evaluation of the fitting and forecasting performance of a set of the most often applied ARIMA models to Canadian economic time series. The question of which models perform well is important for programs like the X-ll-ARIMA (Dagum 1980) which automatically fits a fixed small set of models (three models in the case of the X-ll-ARIMA) to the series"--Introduction, p. 1.

Lien permanent pour cette publication :
publications.gc.ca/pub?id=9.837876&sl=1

• Format MARC XML
• Format MARC HTML

Demander des formats alternatifs