Portfolio optimization using multi-obj ective genetic algorithms
作者： Prisadarng Skolpadungket , Keshav Dahal , Napat Harnpornchai
关键词:
摘要: A portfolio optimisation problem involves allocation of investment to a number different assets maximize yield and minimize risk in given period. The selected not only collectively contribute its but also interactively define as usually measured by variance. In this paper we apply various techniques multiobjective genetic algorithms solve optimization with some realistic constraints, namely cardinality floor constraints round-lot constraints. experimented are Vector Evaluated Genetic Algorithm (VEGA), Fuzzy VEGA, Multiobjective Optimization (MOGA) , Strength Pareto Evolutionary 2nd version (SPEA2) Non-Dominated Sorting (NSGA2). results show that using fuzzy logic combine objectives VEGA (in VEGAFuzl) for does improve performances Generation Distance (GD) defined average distances the last generation population nearest members true front solutions tend cluster around few points. MOGA SPEA2 use diversification they perform better terms finding diverse front. performs best even comparatively small generations. NSGA2 closed GD poor distribution.
sci-hub.se 参考文章(16)
A. Loraschi, Marco Tomassini, A. Tettamanzi, Paolo Verda, Distributed Genetic Algorithms with an Application to Portfolio Selection Problems international conference on adaptive and natural computing algorithms. pp. 384- 387 ,(1995) , 10.1007/978-3-7091-7535-4_100
F. Streichert, H. Ulmer, A. Zell, Evaluating a hybrid encoding and three crossover operators on the constrained portfolio selection problem congress on evolutionary computation. ,vol. 1, pp. 932- 939 ,(2004) , 10.1109/CEC.2004.1330961
A.G.B. Tettamanzi, A. Loraschi, S. Arnone, A genetic approach to portfolio selection Institute of information and computer technology, Czech Polytechnic University. ,vol. 3, pp. 597- 604 ,(1993)
R. Subbu, P.P. Bonissone, N. Eklund, S. Bollapragada, K. Chalermkraivuth, Multiobjective financial portfolio design: a hybrid evolutionary approach congress on evolutionary computation. ,vol. 2, pp. 1722- 1729 ,(2005) , 10.1109/CEC.2005.1554896
Fischer Black, Michael C Jensen, Myron Scholes, The Capital Asset Pricing Model: Some Empirical Tests Social Science Research Network. ,(1972)
Felix Streichert, Andreas Zell, Holger Ulmer, Hybrid Representations for Composition Optimization and Parallelizing MOEAs dagstuhl seminar proceedings. pp. 0- ,(2005)
Dietmar Maringer, Heuristic Optimization for Portfolio Management [Application Notes] IEEE Computational Intelligence Magazine. ,vol. 3, pp. 31- 34 ,(2008) , 10.1109/MCI.2008.929847
Christian Blum, Andrea Roli, Metaheuristics in combinatorial optimization: Overview and conceptual comparison ACM Computing Surveys. ,vol. 35, pp. 268- 308 ,(2003) , 10.1145/937503.937505
E. Zitzler, L. Thiele, Multiobjective evolutionary algorithms: a comparative case study and the strength Pareto approach IEEE Transactions on Evolutionary Computation. ,vol. 3, pp. 257- 271 ,(1999) , 10.1109/4235.797969
Y. Crama, M. Schyns, Simulated annealing for complex portfolio selection problems European Journal of Operational Research. ,vol. 150, pp. 546- 571 ,(2003) , 10.1016/S0377-2217(02)00784-1