Optimal Post-Hoc Theorizing
If "Statistical Learning" exceeds "Darwinian Learning," then it's better to look at the data before theorizing.
High-Throughput Asset Pricing or: How I Learned to Stop Worrying and Love Data Mining
(with Chukwuma Dim)
The solution to data mining bias is to mine data rigorously.
Data for returns of 80,000 long-short strategies based on past returns and ticker symbols
For high-throughput accounting strategies, see "Does peer-reviewed theory help predict the cross-section of stock returns?" (below)
Does Peer-Reviewed Research Help Predict Stock Returns?
(with Alejandro Lopez-Lira and Tom Zimmermann)
What Useful Alphas?
(with Ivo Welch)
If you drop micro caps, focus on post 2005 data, and account for data mining bias, anomaly alphas are gone.
This is basically a distilled version of Chen and Velikov (2022)
Semi-Parametric Restrictions on Production-Based Asset Pricing Models
Matching the data on asset prices requires either extremely volatile IST shocks or huge capital adjustment costs.
The best parts of this paper are extended and can be found in "An Irrelevance Theorem for Risk Aversion and Time-Varying Risk" with Francisco Palomino.