Deep Learning Factor Alpha
日期: 2018-09-12


A deep learning automated solution to generate long-short factors using a high-dimensional firm characteristics. 

Our algorithm performs a nonlinear search and finds the optimal transformation of characteristics used for security sorting, with one asset pricing objective: minimizing alphas.

Our deep factors, hidden neurons in the neural network, are trained greedily with the backward propagation feedback from the loss function that considers both time series and cross-sectional variations. Our conditional forecast generalizes a benchmark, such as CAPM, and includes Fama-French type models as special cases. The conclusion is the improvement of insignificant alphas for some anomalies as well as sorted portfolios.


FENG Guanhao, assistant Professor at the Department of Management Science, College of Business, the City University of Hong Kong. Received a Ph.D. in Econometrics and Statistics from University of Chicago and joined City U as an assistant professor of statistics in 2017.He has worked as a quantitative researcher intern at Citadel for machine learning research.?Gavin’s research interests include financial time series, empirical asset pricing, machine learning, and quantitative finance.?His work on taming the factor zoo in asset pricing earned the 2018 AQR Insight Award. His another work on deep learning asset pricing was awarded by the Unigestion Alternative Risk Premia Research Academy.