A MULTILEVEL MODEL FOR PREDICTING ROAD TRAFFIC FATALITIES IN GHANA

Abstract

A Multilevel Random Coefficient (MRC) model for predicting road traffic fatalities in Ghana is proposed. In this model, the number of road traffic fatalities and the regional groups are conceptualized as a hierarchical system of road traffic fatalities and geographical regions of Ghana, with fatalities and regions defined at separate levels of this hierarchical system. Instead of estimating a separate regression equation for each of the 10 regions in Ghana, a multilevel regression analysis was applied to estimate the values of the regression coefficients for each region based on data given. The result shows that there is significant intercept variation in terms of the dependent variable y across the 10 regions. It was estimated that about 58% of the variation in y is a function of the region to which it is observed, thus, validating the application of the multilevel model. Using the random slope model M2, it was found that, from 2001 to 2012 in Greater Accra region, all the 12 estimated road traffic fatality figures are within 10% of the actual figure. Out of the 22 calculated figures, from 1991 to 2012, 15 are within 10% of the actual figure and 19 are within 20% of the actual value.