Title: Interval and value at risk forecasting for realized volatility and implied volatility having asymmetry
Authors: Ji Eun Choi - Ewha Womens university (Korea, South) [presenting]
Dong Wan Shin - Ewha university (Korea, South)
Abstract: A new strategy is proposed for forecasting the confidence interval (CI) and value at risk (VaR) of Realized volatility (RV) and implied volatility (IV) which fully addresses asymmetry. For all RVs and IVs, significant asymmetries are identified in three parts; mean, volatility and error distribution. No method for asymmetries reflected multi step CI and VaR forecasts are available in the literature. The asymmetries are addressed by the LHAR (Leverage heteroscedastic autoregression) model for mean part, by the EGARCH model for volatility part, by the skew-$t$ distribution for residual part. Considerable out-of-sample forecast improvements of the CI and VaR is demonstrated for three financial assets: the US S$\&$P 500 index, the US NASDAQ index, the Korea KOSPI index. For the RVs and IVs, we get CI forecasts with better coverage with smaller length and better VaR with better violation error if asymmetries are properly considered.