PEMODELAN REGRESI ANTARA KEMALANGAN JALAN RAYA DENGAN ISIPADU TRAFIK, PERSIMPANGAN, LAMPU ISYARAT DAN WAKTU KEJADIAN

This paper discusses a relationship between the numbers of road accidents in one district of the state of Kedah, Malaysia with traffic volume, road junctions, traffic light, and occurring time. Types of road junction that were considered in this study were the T-junctions and four-legged intersections either signalised or non-signalised. The occurring time refers to peak and off-peak time. Likelihood ratio test gives evidence that the Negative Binomial regression model is the best model to fit the accident data. The designs of the regression model refers to accident occurring among each of the traffic volume collection stations in the study district.