Title :
On VEPSO and VEDE for solving a treaty optimization problem
Author :
Carmona Cortes, Omar Andres ; Rau-Chaplin, Andrew ; do Prado, Pedro Felipe
Author_Institution :
Inf. Acad. Dept., Fed. Inst. of Sci., Educ. & Technol. of Maranhao, Sao Luis, Brazil
Abstract :
The purpose of this paper is to evaluate the performance of Vector Evaluated Differential Evolution (VEDE) and Vector Evaluated Particle Swarm Optimization (VEPSO) in solving a real world financial optimization problem. The algorithms have been applied to the Reinsurance Contract Problem, which is a challenging problem in computational finance, and their performance has been evaluated in terms of metrics including the average number of solutions, the average hypervolume and the coverage. Results have shown that both algorithms can reach good solutions, however VEPSO tends to perform better.
Keywords :
evolutionary computation; finance; particle swarm optimisation; vectors; VEDE; VEPSO; average hypervolume; computational finance; real world financial optimization problem; reinsurance contract problem; treaty optimization problem; vector evaluated differential evolution; vector evaluated particle swarm optimization; Analysis of variance; Companies; Equations; Optimization; Sociology; Vectors;
Conference_Titel :
Systems, Man and Cybernetics (SMC), 2014 IEEE International Conference on
Conference_Location :
San Diego, CA
DOI :
10.1109/SMC.2014.6974290