Abstract

The research aims to find volatility models of daily closing price from Iraqi stock market for period (2005 – 2012) using autoregressive conditional heteroscedasticity models (ARCH) when the error distribution is normal (Gaussian) that take into account volatility in prices during periods of circulation , of tests to identify the existence of heteroscedasticity which these models characterized there with. Estimation has been studied and included the using of maximum likelihood estimation method . as well as studying the Diagnostic checking using a number of tests to define the scope of models relevancy that has been estimated for the data examined then forecasting volatility (fluctuations) of prices through volatility forecast daily closing price by using In-sample forecasting method., The results of application on the study data show that the best model to forecast volatility of the daily closing price is GARCH(1,2) and without any effects for ARCH in model, by depend on Akaike Information Criterion (AIC), Schwartz Information Criterion (SIC) , Hannan Quinn Information Criterion (H-Q) , The significance of the estimated parameters of the model, and the accuracy of forecasting by depend on forecasting accuracy criterion (RMSE , MAE , MAPE , Theil inequality coefficient).