A multi-objective continuous genetic algorithm for financial portfolio optimization problem
Yacine Kessaci · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2017
Financial portfolio management with both quantity and cardinality constraints can be modeled as a NP-hard continuous optimization problem where two objectives are optimized: maximizing the return of the portfolio and minimizing its risk. In this paper, we propose a work that aims at developing and tuning a multi-objective continuous genetic algorithm that gives the best Pareto-set of portfolios with different trade-offs between objectives. Experiments have been conducted using realistic pricing history of the CAC40 stock market. The reported results show the configuration of the genetic algorithm with the best performance.