Deep neural expected shortfall regression with tail-robustness

TL;DR

This paper introduces a deep neural network-based expected shortfall regression framework that is robust to heavy-tailed data and addresses the limitations of traditional ES estimation methods, especially in high-dimensional settings.

Contribution

It develops a two-step deep neural network approach for ES regression that is robust to heavy tails and leverages hierarchical structures, filling a gap in existing quantile and ES regression methods.

Findings

The proposed method achieves high accuracy in simulations.

It demonstrates robustness to heavy-tailed distributions.

Empirical analysis shows effectiveness in real-world precipitation data.

Abstract

Expected shortfall (ES), also known as conditional value-at-risk, is a widely recognized risk measure that complements value-at-risk by capturing tail-related risks more effectively. Compared with quantile regression, which has been extensively developed and applied across disciplines, ES regression remains in its early stage, partly because the traditional empirical risk minimization framework is not directly applicable. In this paper, we develop a nonparametric framework for expected shortfall regression based on a two-step approach that treats the conditional quantile function as a nuisance parameter. Leveraging the representational power of deep neural networks, we construct a two-step ES estimator using feedforward ReLU networks, which can alleviate the curse of dimensionality when the underlying functions possess hierarchical composition structures. However, ES estimation is inherently sensitive to heavy-tailed response or error distributions. To address this challenge, we integrate a properly tuned Huber loss into the neural network training, yielding a robust deep ES estimator that is provably resistant to heavy-tailedness in a non-asymptotic sense and first-order insensitive to quantile estimation errors in the first stage. Comprehensive simulation studies and an empirical analysis of the effect of El Ni\~no on extreme precipitation illustrate the accuracy and robustness of the proposed method.