The gene expression messy genetic algorithm for financial applications

Abstract

The paper introduces the gene expression messy genetic algorithm (GEMGA)-a new generation of messy GAs that may find many applications in financial engineering. Unlike other existing blackbox optimization algorithms, GEMGA directly searches for relations among the members of the search space. The GEMGA is an O(|/spl Lambda/|/sup k/(l+k)) sample complexity algorithm for the class of order-k delineable problems (Kargupta, 1995) (problems that can be solved by considering no higher than order-k relations) in sequence representation of length L and alphabet set /spl Lambda/. The GEMGA is designed based on the alternate perspective of natural evolution proposed by the SEARCH framework (Kargupta, 1995) that emphasizes the role of gene expression. The paper also presents the test results for large multimodal problems and identifies possible applications to financial engineering.