Signature-based identification of volatility models from path geometry
Updated
Updated · arxiv.org · Jul 7
Signature-based identification of volatility models from path geometry
1 articles · Updated · arxiv.org · Jul 7
Summary
Researchers have developed a signature-based framework to identify classes of stochastic volatility models directly from observed path data.
By mapping volatility trajectories into a feature space via truncated path signatures and using gradient boosting classifiers, the method distinguishes between models without parametric calibration.
Experiments show high classification accuracy, even under parameter uncertainty, suggesting this approach could offer a robust alternative for model identification in quantitative finance.