Calibration of a path-dependent volatility model: empirical tests

Co-authored with Paolo Foschi. Published in Comput. Statist. Data Anal., Volume 53, pp.2219-2235, 2009

The Hobson and Rogers model for option pricing is considered. This stochastic volatility model preserves the completeness of the market and can potentially reproduce
the observed smile and term structure patterns of implied volatility. A calibration procedure based on ad-hoc numerical schemes for hypoelliptic PDEs is proposed and used to quantitatively investigate the pricing performance of the model. Numerical results based on S&P500 option prices are discussed.