Importance Sampling for Option Greeks with Discontinuous Payoffs

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalNot applicablepeer-review

3 Scopus Citations
View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Original languageEnglish
Pages (from-to)223-235
Journal / PublicationINFORMS Journal on Computing
Volume28
Issue number2
Early online date16 Feb 2016
Publication statusPublished - May 2016

Abstract

The Greeks are derivatives of option price with respect to market parameters and play an important role in financial risk management. Among various simulation methods for estimating the Greeks, the pathwise method typically has a low variance. However, when the option payoff is discontinuous, the pathwise method is not applicable and the Greek involves a conditional expectation taken over a hypersurface that is a probability-zero set. In this paper, we propose an importance sampling (IS) method to estimate this conditional expectation. More specifically, IS is applied in a way that all simulated observations fall into a set constructed by thickening the hypersurface. Allowing the thickness of the set to go to zero then leads to a new representation of the Greek as an ordinary expectation, thus leading to an unbiased estimator. The resulting estimator makes use of the pathwise derivatives and can be viewed as an extension of the pathwise method to cases with discontinuous payoffs. Numerical results show that the proposed IS method works well.

Research Area(s)

  • simulation, importance sampling, risk management, Greek letters