Rough Volatility

Risks & contraindications

Comparisons

Highlighted claims

• MNO outperforms Neural SDE and Neural CDE on rough volatility at all tested Hurst exponents, with the largest margin at H=0.1, the most non-Markovian regime. — Martingale Neural Operators: Learning Stochastic Marginals via Doob-Meyer Factorization
• Targeting the terminal marginal directly without modeling the full non-Markovian path history can outperform sequential baselines optimized for pathwise fidelity on rough volatility tasks. — Martingale Neural Operators: Learning Stochastic Marginals via Doob-Meyer Factorization
• Autoregressive reuse of MNO whitens temporal roughness toward Brownian scaling, marking the boundary of the model's valid scope. — Martingale Neural Operators: Learning Stochastic Marginals via Doob-Meyer Factorization