Agent-Based Co-operative Co-evolutionary Algorithms for Multi-objective Portfolio Optimization

被引：0

作者：

Drezewski, Rafal ^{[1
]}

Obrocki, Krystian ^{[1
]}

Siwik, Leszek ^{[1
]}

机构：

[1] AGH Univ Sci & Technol, Dept Comp Sci, Krakow, Poland

来源：

NATURAL COMPUTING IN COMPUTATIONAL FINANCE, VOL 3 | 2010年 / 293卷

关键词：

MULTIAGENT SYSTEM;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Co-evolutionary techniques makes it possible to apply evolutionary algorithms in the cases when it is not possible to formulate explicit fitness function. In the case of social and economic simulations such techniques provide us tools for modeling interactions between social and economic agents especially when agent-based models of co-evolution are used. In this chapter agent-based versions of multi-objective co-operative co-evolutionary algorithms are presented and applied to portfolio optimization problem. The agent-based algorithms are compared with classical versions of SPEA2 and NSGA2 multi-objective evolutionary algorithms with the use of multi-objective test problems and multi-objective portfolio optimization problem. Presented results show that agent-based algorithms obtain better results in the case of multi-objective test problems, while in the case of portfolio optimization problem results are mixed.

引用

页码：63 / 84

页数：22

共 17 条

[1]

[Anonymous], 1999, Evolutionary Algorithms for Multiobjective Optimization: Methods and Applications

[2]

Deb K., 2010, MULTIOBJECTIVE OPTIM

[3]

DEB K, 2001, SCALABLE TEST PROBLE

[4]

Drezewski R, 2003, LECT NOTES ARTIF INT, V2691, P314

[5]

Drezewski R., 2006, P IEEE WORLD C COMP

[6]

DREZEWSKI R, 2009, NATURAL COMPUTATION, V2

[7]

Drezewski R, 2008, STUD COMPUT INTELL, V100, P271

[8]

Drezewski R, 2008, LECT NOTES ARTIF INT, V5097, P388, DOI 10.1007/978-3-540-69731-2_38

[9]

Drezewski R, 2006, LECT NOTES COMPUT SC, V3993, P871

[10]

Drezewski R, 2006, COMPUT INFORM, V25, P305

← 1 2 →