The basic Monte Carlo procedure for option pricing is appealing and simple to implement, and thus a popular tool for practitioners. Unmodified, though, it suffers from two serious shortcomings: a difficulty in treating early exercise, and only modestly accurate convergence characteristics. The Monte Carlo-based method of handling early exercise features can be improved by generating monotonically varying data that are then amenable to extrapolation procedures, thereby enhancing the overall accuracy and reliability of the scheme. Results for options on one and two underlyings compare extremely favorably with exact results for early exercise options. In a final test of the technique, results are computed for an option on five underlyings.
ASJC Scopus subject areas
- Economics and Econometrics