APPLICATION OF BETA-BINOMIAL AND NEGATIVE BINOMIAL MODELS TO ROAD TRAFFIC ACCIDENTS IN NIGERIA

Abstract

Road traffic accident in Nigeria is known to be a major cause of untimely death. The ultimate goal of this research is to apply Beta Binomial model and Negative Binomial model in tracking the pattern of the roads traffic accident data and this was done based on the secondary data collected from Federal Road Safety Corps, on the fatal, serious and minor road traffic accidents for eight consecutive years. Akaike information criteria (AIC) and Bayesian information criteria (BIC) were used to show the efficiency of Negative Binomial model over Beta Binomial. From the result of the analysis, Negative binomial model performed better in its efficiency and consistent result shown by AIC for the fatal, serious and minor road traffic accidents, it was clear that the Negative Binomial model was the best fit for the data. It was observed from the result that in general the number of fatality in road traffic accidents gradually increases with time as the years go by, the number of people killed by road accidents increases. The kind of vehicle involved in an accident and over speed was found to be associated with the highest factor that influences the rate of road traffic accidents.