摘要

This study examines the forecasting power of the most popular volatility forecasting models in the S&P 500 index market, the Eurodollar futures market, and the 30-year US T-Bond futures market at a daily level using a market-based option-pricing error approach. Comparison has been made between two methods including and excluding implied volatility in option-pricing error approach in forecasting next-day volatilities. To remove any advantage to option-implied volatility, the analysis is performed in two steps. Spurious regression biases and biases in the measurement of volatility forecasts are controlled for. The evidence from this paper supports the use of implied volatility as a proxy for market volatility, as it works best in forecasting next-day realized volatility in all the three US markets. The appropriateness of including implied volatility in option-pricing error approach is also discussed.