Latent-Regime Bias Auditing for Volatility Forecasting

2026-08-02

AI Summary

This study addresses the limitations of conventional volatility forecasting evaluation, which relies on aggregate error metrics and fails to uncover conditional model failures across distinct market regimes. The authors propose the first model-agnostic auditing framework that explicitly incorporates latent market mechanisms into the assessment process. By leveraging time-series representation learning and unsupervised clustering, the framework identifies latent regimes, assigns out-of-sample observations to these regimes, and quantifies regime-specific prediction biases, tail risk underestimation, and associated economic losses. Empirical experiments on cryptocurrency and ETF data reveal that even models exhibiting high overall accuracy suffer from significant regime-dependent biases and severe underestimation of tail risks. These findings underscore the critical need to shift from evaluating average predictive accuracy to analyzing conditional reliability across market states.

Abstract

Volatility forecasts are commonly evaluated with aggregate accuracy metrics such as RMSE and MAE, but these metrics can hide conditional failures that matter for risk management. This paper proposes a model-agnostic audit framework for evaluating whether volatility forecasts remain reliable across latent market regimes. We learn time-series representations of market-state windows, cluster them into regimes using only training information, assign regimes out of sample, and compare aggregate forecast behavior with regime-conditional bias, tail-underprediction, and underprediction-sensitive economic losses. Applied to daily volatility forecasting across cryptocurrency and ETF assets, the audit shows that models with competitive aggregate accuracy can still exhibit substantial regime-specific bias and severe tail underprediction. The results suggest that volatility forecasting should be evaluated not only by average error, but also by where and how forecasts become unreliable. Our framework shifts forecast evaluation from asking which model is most accurate on average to identifying the market regimes in which apparently accurate forecasts fail conditionally. Reproducibility: https://github.com/arthurchagas1/Latent-Regime-Bias-Auditing-for-Volatility-Forecasting

Problem

• volatility forecasting
• latent regimes
• forecast bias
• tail underprediction
• model evaluation

Innovation

• latent regime
• volatility forecasting
• bias auditing
• model-agnostic evaluation
• tail underprediction