Given the rapid growth of financial markets over the past 20 years, along with the explosive development of financial derivatives, an ever-growing need for accurate and efficient volatility forecasting has emerged. Such forecasts have numerous financial applications, such as value-at-risk, hedge ratio, option price and portfolio selection. Recently, the broad availability of intraday trading data has inspired practitioners to investigate their information content in modeling and forecasting the volatility of financial markets. This study aims to propose the introduction of various volatility estimators (overnight volatility (Brooks et al., 2000), PK (Parkinson, 1980), GK (Garman and Klass, 1980), RS (Rogers and Satchell, 1991), RV (Andersen and Bollerslev, 1998), RBP (Barndorff-Nielsen and Shephard, 2004), and VIX) into the conditional variance of GARCH(1,1) model to explore the information value of those estimators for improving out-of-sample volatility forecasts of Nasdaq-100 stock index returns at daily horizon over the period from 2005 to 2013. Empirical results indicate that the inclusion of each volatility estimator considered in this research shows an improvement in the GARCH model with certain degree, except for the overnight volatility (ONV) estimator. In addition, daily ranges (PK, GK, RS) and realized volatilities (RV, RBP) are far more informative than the volatility index (VIX).
Other ID | JA45JB86VN |
---|---|
Journal Section | Research Article |
Authors | |
Publication Date | September 1, 2014 |
Published in Issue | Year 2014 Volume: 4 Issue: 3 |