Single-Step Portfolio Construction through Reinforcement Learning and Economically Interpretable AI
13th January 2021
Joint webinar with INQUIRE Europe
Speaker: Lin William Cong (Cornell University)
Professor Cong and his co-authors propose direct optimisation of investors' objectives via reinforcement learning, an alternative to the widely-adopted two-step portfolio-management paradigm entailing estimating distributions of asset returns.
Building on recent breakthroughs in AI, they develop neural-network-based multi-sequence models tailored to distinguishing features of economic and financial data. The resulting AlphaPortfolio yields stellar out-of-sample performances (e.g., Sharpe ratio above two) that are robust under various economic and trading restrictions. Moreover, they use polynomial-feature-sensitivity and textual-factor analyses to project the model onto linear regression and natural language spaces uncover key market signals, firms' financials, etc., including their rotation and non-linearity, that drive investment performance.
Overall, the authors highlight the utility of reinforcement deep learning in social sciences, especially finance, and provide novel ``economic distillation'' procedures for interpreting AI and big data models.
We are pleased to note that this research was sponsored by an INQUIRE UK Research Grant.
Lin William Cong:
Lin William Cong is the Rudd Family Professor of Management and Associate Professor of Finance at the Johnson Graduate School of Management at Cornell University, where he directs the FinTech Initiative. He is also a Kauffman Junior Fellow, Poets & Quants World Best Business School Professor, advisor to the Wall Street Blockchain Alliance, Luohan Academy Fellow, and serves as editor or associate editor at several leading journals such as the Management Science.
Prior to joining Cornell, he was an assistant professor of Finance at the University of Chicago Booth School of Business where he created courses on ``Quantimental Investment,’’ faculty member at the Center for East Asian Studies, doctoral fellow at the Stanford Institute for Innovation in Developing Economies, and George Shultz Scholar at the Stanford Institute for Economic Policy Research.
Professor Cong’s research spans financial economics, information economics, FinTech and Economic Data Science, and Entrepreneurship (theory and intersection with digitisation and development).
Professor Cong has received numerous accolades such as the AAM-CAMRI-CFA Institute Prize in Asset Management, the CME Best paper Award, Finance Theory Group Best Paper Award, the Shmuel Kandel Award, and has also been invited to speak and teach at hundreds of world-renowned universities, venture funds, technology firms, investment and trading shops, and government agencies such as IMF, Asset Management Association of China, SEC, and federal reserve banks. He received his Ph.D. in Finance and MS in Statistics from Stanford University, and A.M. in Physics jointly with A.B. in Math and Physics from Harvard University.