Road traffic accidents of varying degrees that have occurred in Ghana both in the past and present have resulted in several road fatalities. These has led to the loss of lives, property, broken homes, and as well as leaving behind shattered families and communities, among others despite the indispensable role the road transport sector plays in the economy. This paper sort to explore the most parsimonious and robust linear model for the estimation and forecasting road traffic accidents statistically. This was achieved through the examination of the relationship between road traffic accidents, human and vehicular population in Ghana using linear regression models. Empirical results showed the existence of significant positive relationship between road traffic accidents, vehicle population and human population, with respective correlation coefficients as R1=0.855, R2=0.853, and R3=0.855 indicating a very strong positive association between them. The simple linear regression model between road traffic accidents and vehicle population was adjudged the most robust and parsimonious model based on model diagnostics (residual analysis with plots) coupled with tests of hypothesis. The paper therefore concludes that vehicular population is a very key variable that should not be left out in the policy formulation that would deal with curbing road traffic accidents in Ghana based on available statistics and results.
展开▼