I use machine learning to extract the risk factors identified by firms in their annual
disclosures, quantify firms' exposure to each risk, and construct mimicking portfolios that …

We characterize the structure and origins of missingness for 159 cross-sectional return
predictors and study missing value handling for portfolios constructed using machine …

We introduce a new class of tree-based models, P-Trees, for analyzing (unbalanced) panel
of individual asset returns, generalizing high-dimensional sorting with economic guidance …

We use equity returns to construct a time-varying measure of the interest rate that we call the
zero-beta rate: the expected return of a stock portfolio orthogonal to the stochastic discount …

Our objective is to price the cross-section of asset returns. Despite considering hundreds of
systematic risk factors (“factor zoo”), factor models still have sizable pricing errors. A …

How effectively does the market trade against fund flows to absorb their price impact?
Empirical estimates indicate an inelastic market with high price impacts, in sharp contrast to …

Our objective is to price the cross section of asset returns. Despite considering hundreds of
systematic risk factors (“factor zoo”), factor models still have a sizable pricing error. A …

We propose a sentiment-based shrinkage methodology for estimating the return covariance
matrix in portfolio selection. Our methodology relies on three key insights:(1) the mean …

We study the efficiency gains from asset mispricing through the lenses of an optimal
arbitrage portfolio exploiting a large set of firm characteristics. In particular, we decompose …

Forecasting and Managing Correlation Risks Page 1 Overview Data and Variables Estimation
Methodology Out-of-sample Forecast Performance Applications Robustness Conclusion …