Forecast Car Accident in Saudi Arabia with ARIMA Models
Mahmoud Al-Zyood, King Abdulaziz University, Finance, College of Business, Rabigh, Jeddah, Saudi Araba.
Manuscript received on June 21, 2017. | Revised Manuscript received on June 28, 2017. | Manuscript published on July 05, 2017. | PP: 30-33 | Volume-7 Issue-3, July 2017. | Retrieval Number: C3028077317/2017©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley
©The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/
Abstract: Traffic accidents are the main cause of deaths and injury in Saudi Arabia, this work is a challenge to examine the best ARIMA model for forecast a car accident. Results show that an appropriate model is simply an ARIMA (1, 0, 0, 0) due to the fact that, the ACF has an exponential decay and the PACF has a spike at lag2 which is an indication of the said model. The forecasted car accident cases from 1998 to 2016. The selected model with least AIC value will be selected. We entertained nine tentative ARMA models and Chose that model which has minimum AIC (Akaike Information Criterion).The chosen model is the first one AIC (-0.274306) The selected ARIMA (1, 0) (0, 0), model to forecast for the future values of our time series (car accident). Forecasted for the next 7 years with (95%) prediction intervals The prediction values of traffic accidents show that there will be increasing in deaths and injury coming years.
Keywords: Forecasting, ARIMA models, car accident, Akaike Information Criterion (AIC), Bayessian Information Criterion (BIC).