Hongyu Yao, Zhengyang Chi, Junbin Gao · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.17732 · cited by 0 works (OpenAlex) · open-access PDF
Nous proposons un modèle de volatilité stochastique structuré en temps continu pour la prévision de la variance intégrée d'un jour à l'avance. Le modèle introduit un état de mesure réalisée conjointement évoluant qui alimente directement la dynamique de la variance latente, tandis que les innovations browniennes corrélées capturent la dépendance liée au retour. Les paramètres sont estimés à travers l'apprentissage prédictif basé sur la simulation, et la simulation de Monte Carlo prévoit la variance intégrée. Sur les 17 indices d'actions, le modèle proposé montre une forte performance hors échantillon par rapport à GARCH, GARCH-Itô, Realized GARCH, HAR, et HEAVY sous les critères de référence QLIKE, MSE
Machine translation (Meta M2M100); the English original follows.
We propose a structured continuous-time stochastic volatility model for one-day-ahead integrated-variance forecasting. The model introduces a jointly evolving realized-measure state that feeds directly into latent variance dynamics, while correlated Brownian innovations capture return-linked dependence. Parameters are estimated through simulation-based predictive learning, and Monte Carlo simulation forecasts integrated variance. Across 17 equity indices, the proposed model shows strong out-of-sample performance relative to GARCH, GARCH-Itô, Realized GARCH, HAR, and HEAVY benchmarks under QLIKE, MSE, and MAE. Ablation results show that realized-measure feedback provides substantial incremental forecasting value within the proposed framework, while return-linked Brownian dependence contributes more modest complementary gains. The results support incorporating measurement-driven information directly into continuous-time latent volatility dynamics to improve integrated-variance forecasting.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Stochastic volatility, Volatility (finance), Econometrics, Monte Carlo method, Realized variance, Computer science
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.17732, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange