Fitting the Smile, Smart Parameters for SABR and Heston
Daiwa Capital Markets Europe
Pierre-Yves Henri Rivaille
October 30, 2009
In this paper we revisit the problem of calibrating stochastic volatility models. By finding smart initial parameters, we improve robustness of Levenberg-Marquardt. Applying this technique to the SABR and Heston models reduces calibration time by more than 90% compared to global optimization techniques such as Simplex or Differential Evolution.
Number of Pages in PDF File: 13
Keywords: Stochastic volatility, SABR, Heston, Smile volatility, Calibration, Optimization
Date posted: October 31, 2009