Generative Adversarial Regression (GAR): Learning Conditional Risk Scenarios

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Asadi, Saeed, Li, Jonathan Yu-Meng
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917325283786752
author Asadi, Saeed
Li, Jonathan Yu-Meng
author_facet Asadi, Saeed
Li, Jonathan Yu-Meng
contents We propose Generative Adversarial Regression (GAR), a framework for learning conditional risk scenarios through generators aligned with downstream risk objectives. GAR builds on a regression characterization of conditional risk for elicitable functionals, including quantiles, expectiles, and jointly elicitable pairs. We extend this principle from point prediction to generative modeling by training generators whose policy-induced risk matches that of real data under the same context. To ensure robustness across all policies, GAR adopts a minimax formulation in which an adversarial policy identifies worst-case discrepancies in risk evaluation while the generator adapts to eliminate them. This structure preserves alignment with the risk functional across a broad class of policies rather than a fixed, pre-specified set. We illustrate GAR through a tail-risk instantiation based on jointly elicitable $(\mathrm{VaR}, \mathrm{ES})$ objectives. Experiments on S\&P 500 data show that GAR produces scenarios that better preserve downstream risk than unconditional, econometric, and direct predictive baselines while remaining stable under adversarially selected policies.
format Preprint
id arxiv_https___arxiv_org_abs_2603_08553
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generative Adversarial Regression (GAR): Learning Conditional Risk Scenarios
Asadi, Saeed
Li, Jonathan Yu-Meng
Machine Learning
Optimization and Control
Portfolio Management
Risk Management
We propose Generative Adversarial Regression (GAR), a framework for learning conditional risk scenarios through generators aligned with downstream risk objectives. GAR builds on a regression characterization of conditional risk for elicitable functionals, including quantiles, expectiles, and jointly elicitable pairs. We extend this principle from point prediction to generative modeling by training generators whose policy-induced risk matches that of real data under the same context. To ensure robustness across all policies, GAR adopts a minimax formulation in which an adversarial policy identifies worst-case discrepancies in risk evaluation while the generator adapts to eliminate them. This structure preserves alignment with the risk functional across a broad class of policies rather than a fixed, pre-specified set. We illustrate GAR through a tail-risk instantiation based on jointly elicitable $(\mathrm{VaR}, \mathrm{ES})$ objectives. Experiments on S\&P 500 data show that GAR produces scenarios that better preserve downstream risk than unconditional, econometric, and direct predictive baselines while remaining stable under adversarially selected policies.
title Generative Adversarial Regression (GAR): Learning Conditional Risk Scenarios
topic Machine Learning
Optimization and Control
Portfolio Management
Risk Management
url https://arxiv.org/abs/2603.08553