A Statistical Analysis of Road Accident Fatalities in Malaysia

Abstract

Road fatalities are the number of deaths caused by road accident intentionally or unintentionally. Previous studies show that the number of road fatalities in Malaysia is still concerning and increasing from year to year. This unresolved issue was not only life threatening but it also potentially gives major problems to the economic growth in the country. This study aims to determine the best model in forecasting road accident fatalities in Malaysia. There are two methods used which are Autoregressive Integrated Moving Average (ARIMA) model and Auto-Regressive Poisson model. Each of the model was constructed and analysed by using Statistical Package for Social Science (SPSS) and R Studio, then the models were evaluated and compared to each other based on their error value. The error measure that used in this study are Relative Fit Error (RFE), Mean Absolute Percentage Error (MAPE), Mean Absolute Deviation (MAD), Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) by selecting the lowest value of error measure. The lowest error measure considered as the best model and will be used to forecasts the number of road deaths in five years ahead that is 2021 until 2025. The results indicate that the best model is ARIMA (0,2,1) from the Box-Jenkins method as it has the lowest error measure as compared to Auto-Regressive Poisson model. On this basis, it is recommended that future research to implement ARIMA model in forecasting and to include other type of model to be compared with in measuring the error measure.