出版社：The Japanese Society for Evolutionary Computation
摘要：In this paper, portfolio optimization using loan is formulated as a chance constrained problem in which the money borrowed from a loan is invested in risk assets. Then the chance constrained problem is transformed into a deterministic optimization problem that has an equality constraint. In order to apply conventional Differential Evolution (DE) algorithms to the constrained optimization problem effectively, two types of Genotype-Phenotype (GP) mappings, namely a conventional GP mapping and a newly proposed GP mapping, are compared. As a result of numerical experiments including a two-way analysis of variance (two-way ANOVA), it is shown that the proposed GP mapping outperforms the conventional one because the former enhances the quality of solutions obtained by DE algorithms. By using historical data of assets, an advantage of the investment using loan is also confirmed.
