Box-Jenkins and State Space Model in Forecasting Malaysia Road Accident Cases

Abstract

Abstract Road accident is one of the main causes of death and injury worldwide in this fast-paced modern world. Many developing countries, including Malaysia, are facing serious road accident problems. Forecasting road accident cases has become an important step towards setting the road safety target. Hence, this study aims to develop forecasting models and forecast future trends of monthly road accident cases in Malaysia. The data set on monthly number of accident cases from January 2001 to December 2019 was provided by Polis Diraja Malaysia (PDRM). Box-Jenkins and State space models were developed using the data under study. The models were then evaluated based on in-sample and out-sample evaluation using lowest root mean square error, mean absolute percentage error and mean absolute error. The results show that the basic structural state space model with trend and seasonal component was the most appropriate model in forecasting road accident cases in Malaysia. The 10-year ahead forecast from January 2020 to December 2030 shows that monthly road accident cases in Malaysia have a constant inclining pattern for each year. It is hoped that the finding from this study could become a reference for the authorities of Malaysia in making recommendations in order to improve road safety and reduce road traffic accidents in Malaysia.