We demonstrate that predictable uninformed cash flows forecast market and individual stock returns. Buying pressure from dividend payments (announced weeks prior) predicts higher value-weighted market returns, with returns for the top quintile of payment days four times higher than the lowest....

We theoretically characterize the behavior of machine learning asset pricing models. We prove that expected out-of-sample model performance--in terms of SDF Sharpe ratio and test asset pricing errors--is improving in model parameterization (or "complexity"). Our empirical findings verify the...

We test and compare the effects of introduction of two new financial information technologies, EDGAR and XBRL, on well-known asset pricing anomalies often attributed to mispricing. EDGAR facilitates easier access to public accounting information about public firms; XBRL reduces the cost of...

We examine how sell-side equity analysts strategically disclose information of differing quality to the public versus the buy-side mutual fund managers to whom they are connected. We consider cases in which analysts recommend that the public buys a stock, but some fund managers sell it. We...

Cross-sectional forecasts of conservative and optimistic biases in analyst earnings estimates predict a stock's future returns, especially for firms that are hard to value. Trading strategies--whether based on the component of analyst bias that is correlated with major return anomalies or the...

Economic interactions, such as crowdfunding, often involve sequential actions, observational learning, and contingent project implementation. We incorporate all-or-nothing thresholds in a canonical model of information cascades. Early supporters effectively delegate their decisions to a...