Two Mixed Models to Predict the Volatilities of Stock Prices in Egypt

Abstract

A new mixed model depending on mixing the dynamic conditional correlation model (DCC-GARCH) with a bootstrap mean bias corrected estimator method (BMBCE) is suggested and studied to obtain an efficient model to predict the volatilities of stock prices in Egypt. Moreover, this model is studied the conditional correlation and interactions between variables. It also made a comparison between that model and a Grey GARCH model (1,1) over the period of 26/4/2016 to 22/1/2019. The study found the results of the applied study on EGX30 and EGX70 indices of Egypt stock market showed that the DCC-GARCH-BMBCE model has much better performances in volatility forecasting than the GM-GARCH model. This is proved by analyzing the errors for these models by estimating the difference between the actual values and the estimated values in order to measure the accuracy of the models, involving root mean square error (RMSE) and mean absolute error (MAE).