Forest Through the Trees: Building Cross-Sections of Stock Returns

By Svetlana Bryzgalova, Markus Pelger, Jason Zhu. Social Science Research Network, 2020

84 citations recorded by the source corpus.

We build cross‐sections of asset returns for a given set of characteristics, that is, managed portfolios serving as test assets, as well as building blocks for tradable risk factors. We use decision trees to endogenously group similar stocks together by selecting optimal portfolio splits to span the stochastic discount factor, projected on individual stocks. Our portfolios are interpretable and well diversified, reflecting many characteristics and their interactions. Compared to combinations of dozens (even hundreds) of single/double sorts, as well as machine‐learning prediction‐based portfolios, our cross‐sections are low‐dimensional yet have up to three times higher out‐of‐sample Sharpe ratios and alphas.

Subfields: Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/fbb3f7e26d26154dda82b9a69cf8bb1732309d08

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

1. Luyang Chen
2. Jason Zhu
3. Markus Pelger

In Finance

1. Luyang Chen
2. Jason Zhu
3. Markus Pelger

Citations in the corpus

• Deep Learning in Asset Pricing (2019, methodology): “Bryzgalova, Pelger, and Zhu (2020) also show that mean-variance optimization with regularization can be interpreted as an adversarial approach with parameter uncertainty.”