The Comparison between the BEKK and DVECH Models of Multivariate GARCH Models with Practical Application

Abstract

The Purpose of this research is a comparison between two types of multivariate GARCH models BEKK and DVECH to forecast using financial time series which are the series of daily Iraqi dinar exchange rate with dollar, the global daily of Oil price with dollar and the global daily of gold price with dollar for the period from 01/01/2014 till 01/01/2016.The estimation, testing and forecasting process has been computed through the program RATS. Three time series have been transferred to the three asset returns to get the Stationarity, some tests were conducted including Ljung- Box, Multivariate Q and Multivariate ARCH to Returns Series and Residuals Series for both models with comparison between the estimation and forecasting models based on the criterion, mean Squared error (MSE), compared to the Suitability of these two models of the nature of the data and the ability to Capture the volatility. We concluded that BEKK is better than DVECH in forecasting from the model.