ABSTRACT

Increasing number of road traffic accidents with its resulting deaths and injuries have become an issue of great concern in recent times in Nigeria. Various intervention efforts have been designed to reduce accident rates and traffic mortality level. This study applied autoregressive integrated moving average (ARIMA) time series modelling approach to appraise the trends and pattern of road accidents in Anambra State, Nigeria over 132 months as well as made forecast of road traffic accidents in the State. Out of the various ARIMA models examined, ARIMA(1,1,1) model with the lowest Bayesian Information Criterion, Mean Absolute Percentage Error and Root Mean Square Error, was selected as the most suitable model for the total number of accident cases in the State and ARIMA(0,1,1) model was selected for the fatal accident cases . The autocorrelation and partial autocorrelation plots of the residuals of the models, and the Ljung-Box statistics revealed that the models are free from serial correlation, hence adequate and fit well for accidents data in the State. The forecasts showed a slight decreasing pattern in the total number of accident cases and a slight increasing pattern in the fatal accident cases in Anambra State.This study will help road safety agency in the State to understand the trends and patterns of road traffic accidents in the state and be able to take proper safety measures to curtail accidents occurrences and severity in the state

Keywords: –

ARIMA model, Road traffic accidents, Forecasting, Anambra State, Nigeria

Download full paper