MULTIPLE REGRESSION ANALYSIS OF ROAD TRAFFIC ACCIDENTS: CASE STUDY IN ADDIS ABABA CITY ADMINISTRATION

In this study, the researcher tried to do Multiple Regression Analysis of Road Traffic Accidents: Case Study in Addis Ababa City. Road traffic accidents of varying degrees that have occurred in Addis Ababa both in the past and present have resulted in several road fatalities. These has led to the loss of lives, property and permanent disability. The data for present study has been taken from Central Statistics Agency, Ethiopia Road Authority, Addis Ababa City Road Authority, Transport Authority, Addis Ababa Police Commission and Federal Police Commission. Period: 2008-2009 to 2017-2018. All data are quantitative secondary data. This paper have sort to explore Multiple Linear Regression Model for the estimation and forecasting road traffic accidents statistically using SPSS software version 15. This was achieved through the examination of the relationship between road traffic accidents (death, serious injury, slight injury and property damage) and predictors; driving distance, priority for vehicle, priority for pedestrian, driving by pass, driving Start up and Speed in study area using regression model. Empirical results showed that the existence of significant positive relationship between road traffic accidents (Y1, Y2 and X4) and predictors; X1, X2, X3, X4, X5 and X6 with respective correlation coefficients as R1=0.971, R2=0.978, and R3=0.983 indicating a very strong positive association between them. From the ANOVA table in the output the p-value of model 1, 2 and 4 are less than α = 0.05. Therefore, from the decision rule we fail to accept the null hypothesis and conclude that at least one of the coefficients is not zero and as a result the model is significant. Under the curve fitting estimation, the cubic model was selected for predicting the annual traffic road accident for all categories i.e. (i) Death (ii) Serious injury (iii) Slight injury and (iv) property damage. We conclude that the models thus chosen therefore can be used for estimation and forecasting purposes of road traffic accidents used for the approved responsible bodies of road transport authority. Transport authority traffic management offices and other focal bodies should registered all traffic accident data properly and meaningfully; no missing and omitted data will be exist for the future data base.