Fuzzy logic and genetic algorithms for financial risk management

Abstract

We discuss the applicability of fuzzy logic multi criteria ranking techniques and genetic algorithms in solving problems concerning financial risk management. Fuzzy logic techniques are useful in soliciting information on user perceptions of risk factors. However, since people are notoriously inaccurate and unreliable in reporting their preferences, we also employ a genetic algorithm to help validate user supplied data. The genetic algorithm helps clarify how and when user preferences effect the perceived desirability of a particular outcome. The genetic algorithm also helps tune the parameters of fuzzy multiple criteria decision models.