Optimized Least-squares Monte Carlo for Measuring Counterparty Credit Exposure of American-style Options
Kin Hung (Felix) Kan, Greg Frank, Victor Mozgin, Mark Reesor
Building on the least-squares Monte Carlo (LSM) method that was originally proposed by Longstaff and Schwartz (2001) to price American options, we develop a new version of the LSM method, which we term "optimized least-squares Monte Carlo" (OLSM), to measure the counterparty credit exposure of American options. In order to enhance its performance, OLSM is integrated with the following three techniques: variance reduction, initial state dispersion and multiple bucketing (piecewise linear regression). Numerical results demonstrate the power of the OLSM method.