Predicting Real Economic Activity with Individual Option-Implied Expected Returns

We propose a new predictor to forecast U.S. real economic activity (REA) by utilising the information embedded in equity option prices. We construct our equity option-based predictor by applying standard and recent data reduction methods, to the cross-section of computed option-implied expected returns of the underlying stocks. Our predictor forecasts REA both in- and out-of-sample setting even after controlling for common REA predictors and considering their persistence. We find a robust negative relationship between the option-implied predictor and REA. We show that individual stocks contain some additional predictive power that is not being captured neither by the index option-implied expected return, nor by standard factors.