Jose Blanchet

Jose Blanchet, Stanford

Data-driven Distributionally Robust Markowitz Portfolio Selection via Optimal Transport

(This talk is based on joint work with L. Chen, Y. Kang, K. Murthy, F. Zhang, and X. Y. Zhou)

We introduce a portfolio selection formulation based on classical mean-variance portfolio selection, but with the introduction of an adversarial player, which is allowed to perturb the benchmark model based on historical data. Our formulation can be interpreted as a game in which the manager chooses a portfolio to minimize the variance subject to a target return, while an adversary changes the distribution to maximize the variance while diminishing the return. We motivate our game-theoretic approach by connecting our formulation to popular machine learning estimators and state-of-the-art artificial intelligence training methods. We compare the performance of our estimators with several models used in practice and show that our approach is both easy to implement and exhibits strong empirical performance.

Bio: 

Jose Blanchet is a faculty member in the Management Science and Engineering Department at Stanford University where he also obtained his Ph.D. Prior to joining Stanford he was a faculty member at the departments of IEOR and Statistics at Columbia University and in the Statistics Department at Harvard University. Jose is a recipient of the 2009 Best Publication Award given by the INFORMS Applied Probability Society and of the 2010 Erlang Prize. He also received a PECASE award given by NSF in 2010. He worked as an analyst in Protego Financial Advisors, a leading investment bank in Mexico. He has research interests in applied probability and Monte Carlo methods. He serves in the editorial board of Advances in Applied Probability, Journal of Applied Probability, Mathematics of Operations Research, QUESTA, Stochastic Models, and Stochastic Systems.