Prior theoretical work on estimation risk generally has been restricted to single-period, returns-based models in which the investor must estimate the vector of expected returns but the covariance matrix is known. This paper extends the literature on parameter uncertainty in several ways. First, we analyze asymmetric parameter uncertainty in a model based on payoffs. Second, we explore the effects of both symmetric and asymmetric estimation risk on equilibrium asset prices when the covariance matrix for payoffs must also be estimated. Finally, we investigate the effects on equilibrium of asymmetric parameter uncertainty in a simple multiperiod model.

We empirically evaluate 20 prominent contributions across a broad range of areas in the empirical corporate finance literature. We assemble the necessary data and apply a single, simple econometric method, the connected-groups approach of Abowd et al. to appraise the extent to which prevailing empirical specifications explain variation of the dependent variable, differ in composition of fit arising from various classes of independent variables, and exhibit resistance to omitted variable bias and other endogeneity problems. We assess empirical performance across a wide spectrum of areas in corporate finance and indicate varying research opportunities for empiricists and theorists.