A0405
Title: Tail risk forecasting with semi-parametric regression models by incorporating overnight information
Authors: Cathy W-S Chen - Feng Chia University (Taiwan) [presenting]
Takaaki Koike - Hitotsubashi University (Japan)
Abstract: Realized volatility and overnight information are incorporated into risk models, wherein the overnight return often contributes significantly to the total return volatility. Extending a semi-parametric regression model based on asymmetric Laplace distribution, a family of RES-CAViaR-oc models is proposed by adding overnight return and realized measures as a nowcasting technique for simultaneously forecasting value-at-risk (VaR) and expected shortfall (ES). Bayesian methods are utilized to estimate unknown parameters and forecast VaR and ES jointly for the proposed model family. Extensive backtests are also conducted based on joint elicitability of the pair of VaR and ES during the out-of-sample period. The empirical study on four international stock indices confirms that overnight return and realized volatility are vital in tail risk forecasting.
