Regime-Gated Residual Mixture-of-Experts for Cross-Sectional Volatility Forecasting
2026-08-12 • Machine Learning
Machine Learning
AI summaryⓘ
The authors studied how to best include information about different market conditions (regimes) when using neural networks to predict stock volatility. They created a model called RG-ResMoE that uses regime information only to decide which expert sub-model to activate, rather than including it directly in the prediction input. This approach improved both forecast accuracy and training stability compared to simply adding regime data to the inputs. Their tests on U.S. and Japanese stocks showed that careful use of regime information helps prediction more than just increasing model complexity. They also found that soft routing of experts works better than hard routing in this setup.
financial volatilityregime dependenceneural networksrealized volatilitymixture-of-expertsrouting gateforecasting accuracytraining stabilityValue-at-Risksoft routing
Authors
Junyi Ye, Gargi Vijay Borde
Abstract
Financial volatility is regime dependent, yet incorporating regime information into neural networks can also destabilize training. This paper asks where such information should enter a neural cross-sectional volatility forecasting model. We study five-day realized-volatility forecasts for 1,027 U.S. equities using a rolling walk-forward evaluation framework in which information, model capacity, hyperparameter tuning, and random seeds are matched across architectures. We propose RG-ResMoE, a regime-gated residual mixture-of-experts architecture in which regime information is used only for expert routing rather than for direct forecasting. The base predictor models volatility from stock features, while a gating network uses regime state variables to route residual corrections. RG-ResMoE consistently outperforms a capacity-matched MLP in both forecasting accuracy and training stability in the main U.S. study. Similar gains are observed on an independent Japanese panel. The integration pathway is decisive: appending the same regime variables directly to the forecasting input degrades both predictive performance and training stability, whereas restricting them to the routing gate improves accuracy and Value-at-Risk calibration. Hard routing consistently underperforms soft routing. The results suggest that, in compact neural volatility forecasting models, the primary value of mixture-of-experts models lies less in increasing model capacity than in controlling how nonstationary regime information influences prediction.