Modeling and Forecasting the Algerian Stock Exchange Using the Box-Jenkins Methodology

 Author(s): Mohamed Samir Boudrioua; Abderrahmane Boudrioua 

Abstract: The Algiers Stock Exchange (ASE) is the only stock exchange in Algeria. It’s one of the newest and smallest emerging stock exchanges in the world. The focus of this paper is to model and forecast monthly returns of the ASE index (DZAIRINDEX) using The Box- Jenkins methodology. The period of this study is from June 2010 to May 2020. We split the data into training and testing returns datasets. According to Akaike’s Information Criterion (AIC) estimator, the Seasonal Autoregressive Integrated Moving Average SARIMA (2,0,0)(0,0,1)12 with zero mean is chosen as the best model for forecasting the monthly DZAIRINDEX returns. Diagnostic tests show that the fitted model is adequate, where the residuals of this model are normally distributed with no autocorrelation and no heteroskedasticity. We evaluate our model forecasts on returns testing datasets for 11 steps ahead. We get white noise residuals without heteroskedasticity, which confirm the adequacy of our model. Based on different measures of forecast accuracy such as ME, MAE, RMSE, MASE, we show that the forecast accuracy of SARIMA (2,0,0)(0,0,1)12 is acceptable and this model performs much better than a naïve model. The forecast of the whole returns dataset for one year ahead using this model shows a slight increasing fluctuations trend. These results could be used by the financial communities in Algeria to deal with stock exchange risks and to improve their decisions. 

Keywords: The Algiers Stock Exchange; Forecasting; Box-Jenkins methodology; SARIMA model; Akaike Information Criterion (AIC); Autocorrelation, Heteroskedasticity 
Open Access: This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright: © 2020 Al-Kindi Center for Research and Development

Download Full Paper