Abstract
We evaluate the accuracy of four historical volatility measures and option-implied volatility in predicting future volatility using 10-year daily data for 92 stocks. In addition, we examine the effect of liquidity (proxied by option volume and stock market capitalization) on forecast accuracy. Our results suggest that forecast accuracy using implied volatility compared to historical measures is in general inferior, but for larger firms and higher option trading volume firms implied volatility forecasts are more accurate than historically based forecasts. Among the four historical measures, Parkinson's extreme value method and adjusted mean absolute difference perform best.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
