A portfolio optimization model with three objectives and discrete variables

We formulate the portfolio selection as a tri-objective optimization problem so as to find tradeoffs between risk, return and the number of securities in the portfolio. Furthermore, quantity and class constraints are introduced into the model in order to limit the proportion of the portfolio invested in assets with common characteristics and to avoid very small holdings. Since the proposed portfolio selection model involves mixed integer decision variables and multiple objectives finding the exact efficient frontier may be very hard. Nevertheless, finding a good approximation of the efficient surface which provides the investor with a diverse set of portfolios capturing all possible tradeoffs between the objectives within limited computational time is usually acceptable. We experiment with the current state of the art evolutionary multiobjective optimization techniques, namely the Non-dominated Sorting Genetic Algorithm II (NSGA-II), Pareto Envelope-based Selection Algorithm (PESA) and Strength Pareto Evolutionary Algorithm 2 (SPEA2), for solving the mixed-integer multiobjective optimization problem and provide a performance comparison among them using metrics proposed by the community.

[1]  Panos M. Pardalos,et al.  Handbook of Financial Engineering , 2011 .

[2]  Jih-Jeng Huang,et al.  A novel hybrid model for portfolio selection , 2005, Appl. Math. Comput..

[3]  Jaap Spronk,et al.  The Relevance of Mcdm for Financial Decisions , 2002 .

[4]  Hans Kellerer,et al.  Optimization of cardinality constrained portfolios with a hybrid local search algorithm , 2003, OR Spectr..

[5]  Marco Laumanns,et al.  SPEA2: Improving the strength pareto evolutionary algorithm , 2001 .

[6]  Manfred Gilli,et al.  A Data-Driven Optimization Heuristic for Downside Risk Minimization , 2006 .

[7]  C. B. E. Costa,et al.  A multicriteria model for portfolio management , 2004 .

[8]  Yves Crama,et al.  Simulated annealing for complex portfolio selection problems , 2003, Eur. J. Oper. Res..

[9]  Lothar Thiele,et al.  Multiobjective evolutionary algorithms: a comparative case study and the strength Pareto approach , 1999, IEEE Trans. Evol. Comput..

[10]  Andrea Schaerf,et al.  Local Search Techniques for Constrained Portfolio Selection Problems , 2001, ArXiv.

[11]  Kathrin Klamroth,et al.  An MCDM approach to portfolio optimization , 2004, Eur. J. Oper. Res..

[12]  H. Markowitz.  Portfolio Selection: Efficient Diversification of Investments , 1971 .

[13]  Jonathan E. Fieldsend,et al.  Cardinality Constrained Portfolio Optimisation , 2004, IDEAL.

[14]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[15]  Yue Qi,et al.  Randomly generating portfolio-selection covariance matrices with specified distributional characteristics , 2007, Eur. J. Oper. Res..

[16]  Yue Qi,et al.  Portfolio Selection in the Presence of Multiple Criteria , 2008 .

[17]  Andreas Zell,et al.  Evolutionary Algorithms and the Cardinality Constrained Portfolio Optimization Problem , 2004 .

[18]  Marco Laumanns,et al.  A unified model for multi-objective evolutionary algorithms with elitism , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[19]  Kay Chen Tan,et al.  Evolutionary multi-objective portfolio optimization in practical context , 2008, Int. J. Autom. Comput..

[20]  Carlos A. Coello Coello,et al.  A simple multimembered evolution strategy to solve constrained optimization problems , 2005, IEEE Transactions on Evolutionary Computation.

[21]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems (Genetic and Evolutionary Computation) , 2006 .

[22]  Lothar Thiele,et al.  A Tutorial on the Performance Assessment of Stochastic Multiobjective Optimizers , 2006 .

[23]  Carlos A. Coello Coello,et al.  Evolutionary multi-objective optimization: a historical view of the field , 2006, IEEE Comput. Intell. Mag..

[24]  Martin J. Oates,et al.  The Pareto Envelope-Based Selection Algorithm for Multi-objective Optimisation , 2000, PPSN.

[25]  Piero P. Bonissone,et al.  Multiobjective financial portfolio design: a hybrid evolutionary approach , 2005, 2005 IEEE Congress on Evolutionary Computation.

[26]  G. Mitra,et al.  Computational aspects of alternative portfolio selection models in the presence of discrete asset choice constraints , 2001 .

[27]  G. Mitra,et al.  Mean-risk models using two risk measures: a multi-objective approach , 2007 .

[28]  Marco Laumanns,et al.  Performance assessment of multiobjective optimizers: an analysis and review , 2003, IEEE Trans. Evol. Comput..

[29]  Peter Winker,et al.  New concepts and algorithms for portfolio choice , 1992 .

[30]  José Antonio Lozano,et al.  A multiobjective approach to the portfolio optimization problem , 2005, 2005 IEEE Congress on Evolutionary Computation.

[31]  Gary B. Lamont,et al.  Evolutionary Algorithms for Solving Multi-Objective Problems , 2002, Genetic Algorithms and Evolutionary Computation.

[32]  Yue Qi,et al.  Suitable-portfolio investors, nondominated frontier sensitivity, and the effect of multiple objectives on standard portfolio selection , 2007, Ann. Oper. Res..

[33]  Laura Diosan,et al.  A multi-objective evolutionary approach to the portfolio optimization problem , 2005, International Conference on Computational Intelligence for Modelling, Control and Automation and International Conference on Intelligent Agents, Web Technologies and Internet Commerce (CIMCA-IAWTIC'06).

[34]  Yazid M. Sharaiha,et al.  Heuristics for cardinality constrained portfolio optimisation , 2000, Comput. Oper. Res..

[35]  Carlos A. Coello Coello,et al.  An updated survey of GA-based multiobjective optimization techniques , 2000, CSUR.