Mathematical Modeling of Road Traffic Fatalities
Publication Date: 22/12/2020
Author(s): Matthew Femi Olayiwola.
Volume/Issue: Volume 3 , Issue 5 (2020)
Abstract:
The development of a mathematical model for predicting the future values of road fatality in South Africa is of great concern as road accidents become a public health concern and a threat to the economy. This study aims to develop an ARIMA model for the prediction of quarterly road fatality in all the provinces of South Africa. Nineteen years of quarterly data from 2001 to 2019 were used to build the model. The best two models were selected using the Akaike information criterion (AIC); volatility and adjusted R^2 values. Thus, ARIMA (4,1,1) and ARIMA (4,1,4) were found to be the best models. The ARIMA (4, 1, 4) model forecast produced an excellent match with the actual series. Hence, predicting the future values of road fatality in South Africa using these models will help the policymakers and all stakeholders make well-informed decisions. In the next study, the economic impact of road crashes and fatalities on the South African economy will be considered.
Keywords:
Mathematical Modeling, Road Traffic Fatality, Box-Jerkins, ARIMA, Time Series
