New RG-ResMoE Architecture Improves Financial Volatility Forecasting

A new research paper introduces RG-ResMoE, a novel architecture for forecasting financial volatility by incorporating regime-dependent information. This model uses a gating network to route residual corrections based on regime state variables, rather than directly feeding regime information into the forecasting input. Experiments on U.S. and Japanese equity data show that RG-ResMoE consistently outperforms a capacity-matched multilayer perceptron in both accuracy and training stability, highlighting the importance of how nonstationary regime information influences predictions in neural volatility forecasting models. AI

IMPACT This research could lead to more stable and accurate financial forecasting models by better integrating regime-dependent information.

RANK_REASON The cluster contains a research paper detailing a new neural network architecture for financial forecasting. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• Japan
• multilayer perceptron
• RG-ResMoE
• ScienceCast
• U.S.

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →