AlphaPortfolio: Direct Construction through Deep Reinforcement Learning and Interpretable AI
Abstract/Contents
- Abstract
- We directly optimize the objectives of portfolio management via reinforcement learning—an alternative to conventional supervised-learning-based paradigms that entail first-step estimations of return distributions, pricing kernels, or risk premia. Building upon breakthroughs in AI, we develop multi-sequence neural-network models tailored to the distinguishing features of financial data, while allowing training without labels and potential market interactions. Our AlphaPortfolio yields stellar out-of-sample performances (e.g., Sharpe ratio above two and over 13% risk-adjusted alpha with monthly re-balancing) that are robust under various economic restrictions and market conditions (e.g., exclusion of small stocks and short-selling). Moreover, we project AlphaPortfolio onto simpler modeling spaces (e.g., using polynomial-feature-sensitivity) to uncover key drivers of investment performance, including their rotation and nonlinearity. More generally, we highlight the utility of deep reinforcement learning in finance and “economic distillation” for model interpretation.
Description
Type of resource | text |
---|---|
Date created | July 28, 2021 |
Creators/Contributors
Author | Cong, Lin William |
---|---|
Author | Tang, Ke |
Author | Wang, Jingyuan |
Author | Zhang, Yang |
Organizer of meeting | Judd, Kenneth |
Organizer of meeting | Pohl, Walter |
Organizer of meeting | Schmedders, Karl Schmedders |
Organizer of meeting | Wilms, Ole |
Subjects
Subject | artificial intelligence |
---|---|
Subject | asset pricing |
Subject | explainable AI |
Subject | machine learning, |
Subject | portfolio theory, |
Subject | batched/offline reinforcement learning. |
Genre | Text |
Genre | Working paper |
Genre | Grey literature |
Bibliographic information
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- License
- This work is licensed under a Creative Commons Attribution 4.0 International license (CC BY).
Preferred citation
- Preferred citation
- Cong, L., Tang, K., Wang, J., and Zhang, Y. (2022). AlphaPortfolio: Direct Construction through Deep Reinforcement Learning and Interpretable AI . Stanford Digital Repository. Available at https://purl.stanford.edu/fy908xd8332
Collection
SITE Conference 2021
View other items in this collection in SearchWorksContact information
- Contact
- siteworkshop@stanford.edu
Also listed in
Loading usage metrics...