This paper develops a model in which investors who are prohibited from short selling agree to disagree on the precision of a publicly observed signal. The model implies that the equilibrium price is a convex function of the public signal. The model predicts that (1) the stock price reacts more to good news than to bad news; (2) the skewness of stock returns is positively correlated with contemporaneous returns, but negatively correlated with lagged returns; (3) short sale constraints increase rather than decrease skewness; and (4) disagreement about information precision increases skewness. Empirical tests conducted find supportive evidence for all these predictions.

Risk-neutral and real-world densities are derived from option prices and risk assumptions, and are compared with historical densities obtained from time series. Two parametric risk-transformations are used to convert risk-neutral densities into real-world densities. Both transformations are estimated by maximizing the likelihood of observed index levels, for two parametric density families. Results for the FTSE-100 index show that parametric densities derived from option prices have more explanatory power than historical densities and higher likelihoods than densities estimated by spline methods. A combination of parametric real-world and historical densities provides the preferred predictive densities.