Predicting Penang Road Accidents Influences: Time Series Regression Versus Structural Time Series

Abstract

Road accidents become a major cause of death in recent years. The trends of road accidents keep increasing from year to year. Therefore, this paper will compare two time series methods that are time series regression and structural time series of modeling road accidents in Penang from year 2001 until 2013. There are eight explanatory variables used to model the monthly number of road accidents. In the structural time series, level component is allowed to vary over time, while seasonal components are fixed. Results show that both models have a different view in term of the relationship and the factor contributing to road accidents. However the advantage goes to structural time series method in fitting the best model.