LoadingLoading
- Authors: Shihao Gu, Bryan Kelly, Dacheng Xiu
This book provides an in-depth exploration of the application of machine learning methods in the field of empirical asset pricing, aiming to measure asset risk premiums. The author conducts a comparative analysis of a variety of machine learning techniques, including generalized linear models, dimensionality reduction methods (such as principal component analysis PCA, partial least squares PLS), regularization methods (such as LASSO, Elastic Net), ensemble learning methods (such as gradient boosting regression tree GBRT, random forest) and neural networks (including feedforward neural networks and autoencoders).
This book demonstrates the significant economic benefits for investors using machine learning predictive models, in some cases even doubling the returns of the leading regression-based strategies in the existing literature. The authors identify top-performing methods (such as tree models and neural networks) and attribute their improved predictive power to their ability to capture nonlinear predictor interactions that are missed by traditional methods. Furthermore, the study found that all methods agree on a handful of dominant forecast signals, including variations on momentum, liquidity, and volatility. This book provides a new benchmark for predictive accuracy for machine learning methods that measure aggregate market and individual stock risk premiums, and highlights the growing role of machine learning in financial innovation.
Upgrade to unlock all institutional-grade algorithms, derivative pricing engines, factor backtesting frameworks, and live QuantLab execution.
Explore Membership Access
Pre-order · £490Quantitative analytics platform
Waitlist closed — signal locked. Pre-order TQH TERMINAL now at £490 with launch-day delivery.