A regression-based smoothing spline Monte Carlo algorithm for pricing American options in discrete time |
| |
Authors: | Michael Kohler |
| |
Institution: | (1) Department of Mathematics, Darmstadt University of Technology, Schlossgartenstr. 7, 64289 Darmstadt, Germany |
| |
Abstract: | American options in discrete time can be priced by solving optimal stopping problems. This can be done by computing so-called
continuation values, which we represent as regression functions defined recursively by using the continuation values of the
next time step. We use Monte Carlo to generate data, and then we apply smoothing spline regression estimates to estimate the
continuation values from these data. All parameters of the estimate are chosen data dependent. We present results concerning
consistency and the estimates’ rate of convergence. |
| |
Keywords: | American options Consistency Nonparametric regression Optimal stopping Rate of convergence Regression based Monte Carlo methods Smoothing spline |
本文献已被 SpringerLink 等数据库收录! |