Mining 29,000 accounting ratios for t-statistics over 2.0 leads to cross-sectional predictability similar to the peer review process. For both methods, about 50% of predictability remains after the original sample periods. Data mining generates other features of peer review including the rise in...

We study the determinants of local governments' reliance on bank loans using granular data from the Federal Reserve. Governments that are larger, riskier, rely on historically stable revenue sources, or have higher spending relative to revenues are more likely to borrow from banks. Declines in...

We provide data and code that successfully reproduces nearly all crosssectional stock return predictors. Unlike most metastudies, we carefully examine the original papers to determine whether our predictability tests should produce t-stats above 1.96. For the 180 predictors that were clearly...

We examine recent regulation requiring US municipal governments to disclose private debt. We show that governments fail to disclose 55-80% of reportable debt events and that, conditional on disclosure, filings often omit contract details essential for bond pricing. Non-compliant issuers are also...

We consider parametric portfolio policies of any complexity using deep neural networks to optimize investor utility. Risk aversion acts as an economic regularization mechanism, with higher risk aversion constraining model complexity. Empirically, Deep Parametric Portfolio Policies (DPPP)...