Regression methods for pricing complex American-style options

We introduce and analyze a simulation-based approximate dynamic programming method for pricing complex American-style options, with a possibly high-dimensional underlying state space. We work within a finitely parameterized family of approximate value functions, and introduce a variant of value iteration, adapted to this parametric setting. We also introduce a related method which uses a single (parameterized) value function, which is a function of the time-state pair, as opposed to using a separate (independently parameterized) value function for each time. Our methods involve the evaluation of value functions at a finite set, consisting of "representative" elements of the state space. We show that with an arbitrary choice of this set, the approximation error can grow exponentially with the time horizon (time to expiration). On the other hand, if representative states are chosen by simulating the state process using the underlying risk-neutral probability distribution, then the approximation error remains bounded.

[1]  R. Bellman,et al.  FUNCTIONAL APPROXIMATIONS AND DYNAMIC PROGRAMMING , 1959 .

[2]  Eduardo S. Schwartz,et al.  Convertible Bonds: Valuation and Optimal Strategies for Call and Conversion , 1977 .

[3]  David M. Kreps,et al.  Martingales and arbitrage in multiperiod securities markets , 1979 .

[4]  J. Harrison,et al.  Martingales and stochastic integrals in the theory of continuous trading , 1981 .

[5]  E. V. Doorn,et al.  On the time dependent behaviour of the truncated birth-death process , 1981 .

[6]  I. Karatzas.  On the pricing of American options , 1988 .

[7]  D. Heath,et al.  Bond Pricing and the Term Structure of Interest Rates: A Discrete Time Approximation , 1990, Journal of Financial and Quantitative Analysis.

[8]  Pierre Priouret,et al.  Adaptive Algorithms and Stochastic Approximations , 1990, Applications of Mathematics.

[9]  John Rust.  Using Randomization to Break the Curse of Dimensionality , 1997 .

[10]  John N. Tsitsiklis,et al.  Neuro-dynamic programming: an overview , 1995, Proceedings of 1995 34th IEEE Conference on Decision and Control.

[11]  Jérôme Barraquand,et al.  Numerical Valuation of High Dimensional Multivariate American Securities , 1995, Journal of Financial and Quantitative Analysis.

[12]  John N. Tsitsiklis,et al.  Neuro-Dynamic Programming , 1996, Encyclopedia of Machine Learning.

[13]  John N. Tsitsiklis,et al.  Analysis of Temporal-Diffference Learning with Function Approximation , 1996, NIPS.

[14]  Steven Raymar,et al.  Monte Carlo Estimation of American Call Options on the Maximum of Several Stocks , 1997 .

[15]  P. Glasserman,et al.  Pricing American-style securities using simulation , 1997 .

[16]  P. Glasserman,et al.  Enhanced Monte Carlo Estimates for American Option Prices , 1997 .

[17]  Benjamin Van Roy.  Learning and value function approximation in complex decision processes , 1998 .

[18]  John N. Tsitsiklis,et al.  Optimal stopping of Markov processes: Hilbert space theory, approximation algorithms, and an application to pricing high-dimensional financial derivatives , 1999, IEEE Trans. Autom. Control..

[19]  Francis A. Longstaff,et al.  Valuing American Options by Simulation: A Simple Least-Squares Approach , 2001 .

[20]  James A. Tilley.  Valuing American Options in a Path Simulation Model , 2002 .

[21]  San-Lin Chung,et al.  Pricing American Options Using Monte Carlo Simulation , 2002 .