Forecasting downside and upside realized volatility: The role of asymmetric information

Abstract

This study examines which asymmetric variables lead to the better forecast performance of downside and upside risks. The models used in this study measure downside and upside risks using realized semivariance. In addition to their past values, the models utilize return, volume, and jump components as asymmetric variables. We apply these models to major exchange-traded funds (ETFs) and show that asymmetric return variables increase the forecast performance of downside and upside risks for all ETFs. For bond, commodity, and crude oil ETFs, asymmetric trading volume variables are also found to be an important factor in better forecast performance. These results indicate that asymmetric information plays an important role in forecasting downside and upside risks, enabling superior risk management and investment strategy formulation.