arXiv q-fin· Damiano Brigo, Rapha\"el Huser, Dan Leonte·· 5 hr agoAI Score40
Uncertainty and Interpretability in Deep Rough Volatility: A Neural Information-Theoretic Posterior Approach
Uncertainty and Explainability in Deep Rough Volatility: A Neural Information-Theoretic Posterior Approach
AI summary
For implied volatility surface calibration of the rough Heston (rHeston) model, a simulation-based inference framework is proposed to learn the posterior distribution of model parameters, combining neural ratio estimation with a heteroscedastic neural surrogate pricer to generate uncertainty-aware option price intervals.
Source:arXiv q-fin · arxiv.org