USI - Email
Empirical Asset Pricing via Gaussian Process Regression
Host: Prof. Paul Schneider
USI Campus EST, Room D1.14, Sector D
16:30 - 17:30
Ecole Polytechnique Fédérale de Lausanne (EPFL)
Click here to join
Abstract: We introduce an ensemble learning method based on Gaussian Process Regression (GPR) for predicting expected stock returns conditional on stock-level and macro-economic features. Our ensemble learning approach significantly reduces the computational complexity inherent in GPR inference. We perform an empirical analysis on a large cross-section of US stocks from 1962 to 2016. We find that our method dominates existing machine learning models both statistically and economically, in terms of out-of-sample R2 and Sharpe ratios of prediction-sorted portfolios, respectively. Exploiting the Bayesian nature of GPR, we also construct the optimal mean-variance and minimum variance portfolios based on the posterior covariance matrix. Both perform significantly better than the S&P 500 and equally-weighted portfolio.
Biography: Damir Filipović holds the Swissquote Chair in Quantitative Finance at the Ecole Polytechnique Fédérale de Lausanne (EPFL) and a Swiss Finance Institute Senior Chair. He also acts as head of the Swiss Finance Institute @ EPFL. He holds a Ph.D. in mathematics from ETH Zurich and has been a faculty member of the University of Vienna, the University of Munich and Princeton University. He also worked for the Swiss Federal Office of Private Insurance as co-developer of the Swiss Solvency Test. He is on the editorial board of several academic journals. His research focus is in quantitative finance and risk management. His papers have been published in a variety of academic journals including the Journal of Finance, Journal of Financial Economics, Mathematical Finance, Finance and Stochastics, and the Annals of Applied Probability. He is the author of a textbook titled Term-Structure Models.
INF Newsletter Archive