|
|
|
AUTHOR(S): Awa Terence Achiri, Azeh Gerald, Innocent Ndoh Mbue
|
|
TITLE Prediction of the Mortality Caused by Road Traffic Injury in Cameroon |
 PDF |
|
ABSTRACT Road traffic injuries and deaths are a major global public health problem. A time series model of these phenomena could help decision makers to identify at-risk populations, and predict the likelihood of future events. Therefore, this study aims to predict the mortality caused by road traffic injury in Cameroon from 2000 to 2021using a seasonal autoregressive integrated moving average (SARIMA) model. The database was created by triangulating the country's road safety profile from the World Bank Group's website with local data. Using the Augmented Dickey-Fuller test, the series after differencing and log transformation was found to be stationary (p<.05). The built models were evaluated using the Bayesian information criterion, Root Mean Squared Error, Mean Absolute Error, and Mean Absolute Percentage Error. Using the autocorrelation function (ACF) and partial autocorrelation function (FACF) of residuals and the LjungeBox test, the goodness-of-fit of the various models was compared. Using the verified SARIMA model, four-year mortality projections were made (2022-2025). Based on minimal diagnostic statistics (RMSE = 1.375, MAE =0.965, MAPE = 37.69), the SARIMA (3, 1, 3)x(0, 1, 2)12 model was deemed to be the most parsimonious model from the above evaluations. At the 5% significance level, a Box-Ljung test indicated that there was no white noise (〖x^2〗_((df=10))=5.148; p = .881). In addition, neither ACF nor PACF plots exhibited any peaks outside the negligible region. The projected death rate exhibits an upward tendency over time. There is an urgent need for policymakers to ensure safe, cheap, accessible, and sustainable transport systems for all, as well as the enhancement of road safety, particularly through the expansion of public transit, with a focus on the needs of those in vulnerable situations. |
|
KEYWORDS Augmented Dickey-Fuller test, Autocorrelation functions, Box-Ljung test, Road traffic injury, Mortality, SARIMA |
|
|
|
Cite this paper Awa Terence Achiri, Azeh Gerald, Innocent Ndoh Mbue. (2022) Prediction of the Mortality Caused by Road Traffic Injury in Cameroon. International Journal of Transportation Systems, 7, 46-57 |
|
|
