Volume 18, Special Issue on Computing Technology and Information Management, 2021
Use GARCH Models to Build a Econometric Model to Predict Average Daily Closing Prices of the Iraqi Stock Exchange for the Period 2013-2016
Dr. Abed Ali Hamad and Dr. Ahmad Hussein Battal
Abstract
This research aims to build a standard model for the analysis and prediction of the average daily closing price fluctuations for companies registered in the Iraq Stock Exchange for the period 07/01/2013 to 30/06/2016, using the conditional generalized Heteroscedasticity Generalized Autoregressive (GARCH) models. As these models deal with the fluctuations that occur in the financial time series. The results of the analysis showed that the best model for predicting the volatility of average closing prices in the Iraq Stock Exchange is the EGARCH model (3,1), depending on the statistical criteria used in the preference between the models (Akaike Information Criterion, Schwarz Criterion), and these models can provide information for investors in order to reduce the risk resulting from fluctuations in stock prices in the Iraqi financial market.
Pages: 385-400
DOI: 10.14704/WEB/V18SI04/WEB18136
Keywords: Conditional Variance, Return, Akaike Information Criterion, Autoregressive Conditional Heteroskedastic (ARCH), Mean Absolute Error.