Forecasting the use of Generalized Autoregressive Conditional Heteroscedastic Models (GARCH) Seasonality with practical application

Abstract

In this paper  has been one study of autoregressive generalized conditional heteroscedasticity models existence of the seasonal component, for the purpose applied to the daily financial data at high frequency is characterized by Heteroscedasticity seasonal conditional, it has been depending on Multiplicative seasonal Generalized Autoregressive Conditional Heteroscedastic Models Which is symbolized by the Acronym (SGARCH) , which has proven effective expression of seasonal phenomenon as opposed to the usual GARCH models. The summarizing of the research work studying the daily data for the price of the dinar exchange rate against the dollar, has been used autocorrelation function to detect seasonal first, then was diagnosed with a problem of heteroscdastic , passing through the phase estimation using the method of Maximum Likelihood Conditional and on the assumption that the random error is distributed normal distribution with the application on more than one rank for seasonal model, then determine the appropriate rank of the specimen using a variety of standards down to the prediction phase, it has been shown through the application on the study data stages that the best model for predicting volatility is  SGARCH (1,0)(1,0).