Empirical Asset Pricing via Ensemble Gaussian Process Regression

2022-12-02

Social Science Research Network

AI Summary

This paper addresses the need of low-risk-averse investors for robust return forecasts and portfolio construction. Method: We propose a Gaussian Process Regression (GPR)-based ensemble framework tailored for asset pricing, integrating stock-specific features and macroeconomic variables; incorporating Bayesian uncertainty quantification; designing a lightweight online learning algorithm to substantially reduce GPR’s computational complexity; and constructing mean-variance optimal portfolios grounded in the predictive uncertainty distribution. Contribution/Results: To our knowledge, this is the first systematic application of GPR to cross-sectional stock return prediction and the first introduction of an uncertainty-aware portfolio optimization paradigm. Empirical analysis on U.S. equity data from 1962–2016 demonstrates that our model achieves significantly higher out-of-sample R² and Sharpe ratio than leading machine learning benchmarks; moreover, uncertainty-weighted portfolios consistently outperform the S&P 500 index.

Abstract

We introduce an ensemble learning method based on Gaussian Process Regression (GPR) for predicting conditional expected stock returns given stock-level and macro-economic information. Our ensemble learning approach significantly reduces the computational complexity inherent in GPR inference and lends itself to general online learning tasks. We conduct 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 statistically and economically in terms of out-of-sample $R$-squared and Sharpe ratio of prediction-sorted portfolios. Exploiting the Bayesian nature of GPR, we introduce the mean-variance optimal portfolio with respect to the prediction uncertainty distribution of the expected stock returns. It appeals to an uncertainty averse investor and significantly dominates the equal- and value-weighted prediction-sorted portfolios, which outperform the S&P 500.

Problem

• Stock Return Prediction
• Portfolio Optimization
• Risk Aversion

Innovation

• Gaussian Process Regression
• Predictive Uncertainty
• Optimal Portfolio Strategy