Abstract:
In this paper we explored the relevance of asymmetry and long memory in modeling and forecasting the conditional volatility and market risk of equity market in Iran capital Market (Tehran Stock exchange(TSE) and Iran Fara Bourse(IFB)). A broad set of the most popular linear and nonlinear GARCH (generalized autoregressive conditional Heteroskedasticity)-type models is used to investigate this relevancy of asymmetry and long memory. Our in sample and out-of-sample results displayed that volatility of commodity returns can be better described by nonlinear volatility models accommodating the long memory and asymmetry features. In particular, the FIAPARCH (Fractionally Integrated Asymmetric Power ARCH) model is found to be the best suited for estimating the VaR forecasts for both short and long trading positions. This model given a risk exposure at the 99% confidence interval level have Several implications for equity market risks, policy regulations and hedging strategies can be drawn from the obtained results of this paper.
Machine summary:
The results of this research show that considering asymmetric effects in return series as well as long-term memory effects leads to improving the volatility forecasting and Value at Risk of these two time series.
Mike So2 (2006) and Philip Yu3 (2006), by examining the performance of various GARCH models in estimating Value at Risk using data related to 12 major stock indices in Europe, North America, and Southeast Asia, and four major currencies, found that asymmetric behavior is observed in data related to financial markets, while symmetric behavior is observed in exchange rate returns.
They found long-term memory effects among their used time series and, using models from the GARCH family, found that the FIAPARCH model has the best performance among the other models used for volatility prediction and Value at Risk estimation of these commodities.
Thus, it can be stated that the best parametric model for forec 135 asting volatility in the returns of the two markets, Tehran Stock Exchange and Iran Fara Bourse, are conditional heteroskedasticity models considering asymmetric effects and long-term memory.
/ 6-4- Estimating and Evaluating Value at Risk Using Conditional Heteroskedasticity Models The main objective of this research, namely estimating the Value at Risk (VaR) of the return portfolio of the Tehran Stock Exchange and Iran Fara Bourse index while considering long-term memory effects.
Thus, it can be stated that the best parametric model for predicting volatility in the returns of the two markets, Tehran Stock Exchange and Iran Fara Bourse, is the conditional heteroscedasticity models considering asymmetric effects and long-term memory.
(
, 
, 
)
