The Uncertainty of Machine Learning Predictions in Asset Pricing

Yuan Liao; Xinjie Ma; Andreas Neuhierl; Linda Schilling

doi:10.48550/arxiv.2503.00549

Back

The Uncertainty of Machine Learning Predictions in Asset Pricing

Preprint

Open access

The Uncertainty of Machine Learning Predictions in Asset Pricing

Yuan Liao, Xinjie Ma, Andreas Neuhierl and Linda Schilling

ArXiv.org

Cornell University

03/01/2025

DOI: 10.48550/arxiv.2503.00549

Files and links (1)

url

https://doi.org/10.48550/arxiv.2503.00549View

Preprint (Author's original)This preprint has not been evaluated by subject experts through peer review. Preprints may undergo extensive changes and/or become peer-reviewed journal articles. Open Access

Abstract

Machine learning in asset pricing typically predicts expected returns as point estimates, ignoring uncertainty. We develop new methods to construct forecast confidence intervals for expected returns obtained from neural networks. We show that neural network forecasts of expected returns share the same asymptotic distribution as classic nonparametric methods, enabling a closed-form expression for their standard errors. We also propose a computationally feasible bootstrap to obtain the asymptotic distribution. We incorporate these forecast confidence intervals into an uncertainty-averse investment framework. This provides an economic rationale for shrinkage implementations of portfolio selection. Empirically, our methods improve out-of-sample performance.

Statistics - Machine Learning

Details

Title: Subtitle: The Uncertainty of Machine Learning Predictions in Asset Pricing
Creators: Yuan Liao
Xinjie Ma
Andreas Neuhierl
Linda Schilling
Resource Type: Preprint
Publication Details: ArXiv.org
DOI: 10.48550/arxiv.2503.00549
ISSN: 2331-8422
Publisher: Cornell University; Ithaca, New York
Language: English
Date posted: 03/01/2025
Academic Unit: Economics; Accounting
Record Identifier: 9984937788402771

Metrics

36 Record Views