On modelling seasonal ARIMA series: Comparison, application and forecast (number of injured in road accidents in northeast Algeria)

Abstract

This paper study and modeless a number of road accidental injuries in the region of Skikda (northeast Algeria) according to Box-Jenkins method using EViews software using series data from January 2001 to December 2016. Also, Kalman filter method is given. To this end, Kalman filter method is used for short term prediction and parametric identification purpose. The other side, a comparative study is given to compare between the two methods by de following criteria: Mean absolute percentage error (MAPE), root mean square percentage error (RMSPE) and the Theils’s U statistic. This application used Eviews 5.0 and SPSS 26 software’s.