The Informational Content of Implied Volatility in Individual Stocks and the Market∗†

We examine the informational content of historical and implied measures of variance through an evaluation of forecasts over horizons ranging from 1 to 22 days. These forecasts use heterogeneous autoregressive (HAR) regressions which are constructed with high-frequency data. Our results show that the t and forecasting ability of models based on historical realized variance (RV) increases with the addition of implied volatility in the regression model. We nd that robust regression is better than OLS in forecasting RV outside of the estimation sample. The paper evaluates data from individual equities and the S&P 500. 1 I would like to thank Professors George Tauchen, Tim Bollerslev, Bjorn Eraker and Victor Todorov for their advice and encouragement. I would also like to thank Tzuo Hann Law for helping me obtain the data for this project. My peers, Peter van Tassel and Warren Davis, in the Honors Workshop provided great support. I would not have been able to write such an in-depth thesis without nancial assistance from the Davies Fellowship.