Volatility Modeling and Forecasting of the Egyptian: Stock Market Index using ARCH Models

Abstract

This paper estimates and evaluates the forecasting performance of four alternative ARCH- type Models for predicting stock price index volatility using daily Egyptian data. The competing Models include GARCH, EGARCH, GJR and APAPCH used with four different distributions, Gaussian normal, Student-t, Generalized Error Distribution and skewed Student–t. The estimation results show that the forecasting performance of asymmetric GARCH Models (GJR and APARCH),especially when fat-tailed asymmetric densities are taken into account in the conditional volatility, is better than symmetric GARCH. Moreover, it is found that the APAPCH (1,1) Model Provides the best out-of-sample forecasts among all the candidate Models and the skewed Student-t density is more appropriate for modeling the Egyptian stock market index volatility.