【主讲】周雨田 (中央研究院)
【主题】Forecasting Volatility with Multiple Horizon Extreme Values
【时间】2015年11月19日 (周四) 16：00-17：30
【地点】上海财经大学经济学院楼701室
【语言】英文
【摘要】This paper exploits extreme asset prices (high/low range) of multiple horizons because the long horizon range helps to preserve the persistence inherent in volatility shocks and then improves the predictive ability of future volatility. We develop a range-based multiple horizon volatility model to characterize the dynamics of asset prices in the first two moments: the Multiple Horizon Conditional Autoregressive Range Model (henceforth MHCARR). The model is used for forecasting in comparison with the standard CARR model of Chou (2005). Our empirical analysis shows that the MHCARR model performs significantly better than the standard CARR model both in terms of in-sample fit and out-of-sample forecasting ability for the S&P 500 index.
