Forward indifference valuation of American options

Tim Leung, Ronnie Sircar, Thaleia Zariphopoulou

Research output: Contribution to journalArticlepeer-review

12 Scopus citations


We analyse the valuation of American options under the forward performance criterion introduced by Musiela and Zariphopoulou [Quant. Finance 9 (2008), pp. 161-170]. In this framework, the performance criterion evolves forward in time without reference to a specific future time horizon, and may depend on the stochastic market conditions. We examine two applications: the valuation of American options with stochastic volatility and the modelling of early exercises of American-style employee stock options. We work with the assumption that forward indifference prices have sufficient regularity to be solutions of variational inequalities, and provide a comparative analysis between the classical and forward indifference valuation approaches. In the case of exponential forward performance, we derive a duality formula for the forward indifference price. Furthermore, we study the marginal forward performance price, which is related to the classical marginal utility price introduced by Davis (Mathematics of Derivatives Securities, Cambridge University Press, 1997, pp. 227-254). We prove that, under arbitrary time-monotone forward performance criteria, the marginal forward indifference price of any claim is always independent of the investor's wealth and is represented as the expected discounted pay-off under the minimal martingale measure.

Original languageEnglish (US)
Pages (from-to)741-770
Number of pages30
Issue number5-6
StatePublished - Oct 2012

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Modeling and Simulation


  • American options
  • employee stock options
  • forward performance
  • indifference pricing
  • marginal forward performance price
  • stochastic control with optimal stopping


Dive into the research topics of 'Forward indifference valuation of American options'. Together they form a unique fingerprint.

Cite this