Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5102791 | Physica A: Statistical Mechanics and its Applications | 2017 | 7 Pages |
Abstract
In this study, we investigate the predictability of the realized skewness (RSK) and realized kurtosis (RKU) to stock market volatility, that has not been addressed in the existing studies. Out-of-sample results show that RSK, which can significantly improve forecast accuracy in mid- and long-term, is more powerful than RKU in forecasting volatility. Whereas these variables are useless in short-term forecasting. Furthermore, we employ the realized kernel (RK) for the robustness analysis and the conclusions are consistent with the RV measures. Our results are of great importance for portfolio allocation and financial risk management.
Related Topics
Physical Sciences and Engineering
Mathematics
Mathematical Physics
Authors
Dexiang Mei, Jing Liu, Feng Ma, Wang Chen,