Forecasting Stock Market Volatility Using Implied Volatility

We explored the firm-level forecasting power of implied volatility on realized volatility over various horizons. All existing literatures focused on examining forecasting power over the remaining life of options. We built a linear regression model using implied volatility series to forecast future volatility of various horizons. We compared the result with some historical methods and found that the linear regression implied volatility model compares favorably with the moving average method and with GARCH (1,1) for forecasting future volatility over various forecast horizons both in-the-sample and out-of-sample. In addition, we examined whether implied volatility of equity index options is useful in providing volatility information of a firm. This is necessary since not all companies have options listed and traded in an exchange. Finally, we documented that the forecasting power of implied volatility is related to volume ratio-option trading volume versus stock trading volume. Our evidence indicates that a highly liquid option market is necessary for implied volatility to incorporate all relevant information about future volatility.