An Evaluation of Some Generalized Auto Regressive Conditional Heteroscedasticity Models (GARCH) "An Econometrics Study"

Document Type : Original Article

Authors

Statistics,Mathematics and Insurance Department Faculty of Commerce Alexandria University Alexandria Egypt

Abstract

The contributions of error distributions have been ignored while modeling stock market volatility in Egypt and studies have shown that the application of appropriate error distribution in volatility model enhances efficiency of the model. Using some listed companyies in the Stock Egyptian Exchange, this study estimates symmetric and asymmetric volatility models each in Normal, Student’st and generalized error distributions with the view to selecting the best forecasting volatility model with themost appropriate error distribution. The results suggest the presence of leverage effect meaning that volatility responds more to bad news than it does to equal magnitude of good news in some cases. The last Thirty days out of sample forecast adjudged GARCH and Exponential GARCH models as the best predictive model based on Mean Square Error. The study therefore recommends that empirical works sho-uld consider alternative error distributions with a view to achieveng a robust volatility forecasting model.

Keywords

Files

Share

How to cite

Statistics