Modeling Returns and Volatilities by Asymmetric GARCH Models

A comprehensive empirical analysis of the return and conditional variance of Tehran Stock exchange (TSE) index time series is performed using GARCH models. The performance of these conditional changing variance models are compared to asymmetric EGARCH, TGARCH and PGARCH models. Our main goal is to study the presence of stylized facts and other features in TSE index. To estimate the parameters of the models Maximum Likelihood estimation (MLE) method is applied. Model fit measures AIC, BIC and likelihood is calculated for each fitted model. Our results show that so-called stylized facts such as volatility clustering, leptokurtosis and leverage effect is present in TSE index returns and moreover, the PGARCH model using a Student-t distribution is the most successful in fitting the TSE index returns.

کلیدواژه ها:

asymmetry ، volatility ، GARCH ، TSE ، Student-t