Fuzzy Real Options for Risky Project Evaluation Using Least Squares Monte-Carlo Simulation

A numerical technique for evaluating risky projects with fuzzy real options is developed. Fuzzy real options are based on hybrid variables that represent the market risk of a project, which is derived from data, and the private risk, which is usually estimated by experts. These hybrid variables can be evaluated using an extension of Least Squares Monte-Carlo simulation that produces numerical evaluations of fuzzy real options based on the generation and backward induction of sample paths. A major advantage of this methodology is its ability to determine values regardless of whether or not an analytic solution exists. To illustrate, two fuzzy real options models are evaluated using the proposed algorithm: one, on brownfields, for comparison with analytic outputs for fuzzy real options; the other, on oil development, for comparison to the results of the Integrated Valuation Procedure (IVP), another algorithm to assess private risk. The results indicate that the generalized Least Squares Monte-Carlo simulation produces similar results to the analytic valuation of fuzzy real options, when this is possible. Moreover, the use of fuzzy real options can overcome the private risk problem without invoking IVP, which is preferable because expert linguistic estimates are easier to use in a fuzzy environment.