Title:Online Local Volatility Calibration by Convex Regularization with Morozov's Principle and Convergence Rates

Abstract:In this article we address the regularization of the ill-posed problem of determining the local volatility surface (as a function of time to maturity and price) from market given option prices. We integrate the ever-increasing flow of option price information into the well-accepted local volatility model of Dupire. This leads to considering both the local volatility surfaces and their corresponding prices as indexed by the observed underlying stock price as time goes by in appropriate function spaces.
The parameter to data map consists of a nonlinear operator that maps the (variable) diffusion coefficient of a parabolic initial value problem into its solutions evaluated at certain sets. We tackle the inverse problem by convex regularization techniques in appropriate Bochner-Sobolev spaces.
As a preparation, we prove key regularity properties that enable us to apply convex regularization techniques. This forward framework is then used to build a calibration technique that combines online methods with convex Tikhonov regularization tools. Such procedure is used to solve the inverse problem of local volatility identification. As a result, we prove convergence rates with respect to noise and a corresponding Morozov discrepancy principle for the regularization parameter. We conclude by illustrating and validating the theoretical results by means of numerical tests with synthetic as well as real data.