A novel hybrid algorithm based on a harmony search and artificial bee colony for solving a portfolio optimization problem using a mean-semi variance approach

Abstract

Portfolio selection is one of the major capital allocation and budgeting issues in financial management, and a variety of models have been presented for optimal selection. Semi-variance is usually considered as a risk factor in drawing up an efficient frontier and the optimal portfolio. Since semi-variance offers a better estimation of the actual risk portfolio, it was used as a measure to approximate the risk of investment in this work. The optimal portfolio selection is one of the non-deterministic polynomial (NP)-hard problems that have not been presented in an exact algorithm, which can solve this problem in a polynomial time. Meta-heuristic algorithms are usually used to solve such problems. A novel hybrid harmony search and artificial bee colony algorithm and its application were introduced in order to draw efficient frontier portfolios. Computational results show that this algorithm is more successful than the harmony search method and genetic algorithm. In addition, it is more accurate in finding optimal solutions at all levels of risk and return.

Keywords

portfolio optimizations / mean-variance model / mean semi-variance model / harmony search and artificial bee colony / efficient frontier

References

[1] GeemZ W, KimJ H, LoganathanG V. A new heuristic optimization algorithm: Harmony search [J]. Simulations, 2001, 76: 60-68

[2] MahdaviM, FesangharyM, DamangirE. An improved harmony search algorithm for solving optimization problems [J]. Appl Math Comput, 2007, 188: 1567-1579

[3] LeeK S, GeemZ W, LeeS H B K-W. The harmony search heuristic algorithm for discrete structural optimization [J]. Eng Optim, 2005, 37: 663-684

[4] GeemZ W. Optimal cost design of water distribution networks using harmony search [J]. Eng Optim, 2006, 38: 259-280

[5] GeemZ W. Harmony search algorithm for solving Sudoku [C]//. Proceedings 11th International Conference, 2007HeidelbergSpringer371-378

[6] ForsatiR, HaghighatA T, MahdaviM. Harmony search based algorithms for bandwidth-delay-constrained least-cost multicast routing [J]. Comput Commun, 2008, 31: 2505-2519

[7] CylanH, HaldenbilenS, HaldenbilenS, BaskanO. Transport energy modeling with meta-heuristic harmony search algorithm, an application to Turkey [J]. Energy Policy, 20082527-2535

[8] SakaM P, ErdalF. Harmony search based algorithm for the optimum design of grillage systems to LRFD-AISC [J]. Struct Multidiscip Optim, 2009, 38: 25-41

[9] WuB, QianC, NiW, FanS. Hybrid harmony search and artificial bee colony algorithm for global optimization problems [J]. Computers & Mathematics with Applications, 2012, 64(8): 2621-2634

[10] MarkowitzH. Portfolio Selection [J]. Journal of Finance, 1952, 7(1): 77-91

[11] XueQ, WangZ, LiuS, ZhaoD. An improved portfolio optimization model for oil and gas investment selection [J]. Pet Sci, 2014, 11: 181-188

[12] JiaJ, DyerJ S. A standard measure of risk and risk-value models [J]. Management Science, 1996, 42(12): 1691-1705

[13] FernandezA, GomezS. Portfolio selection using neural networks [J]. Computers & Operations Research, 2007, 34: 1177-1191

[14] XiaY, LiuB, WangS, LaiK. A model for portfolio selection with order of expected returns [J]. Computers and Operations Research, 2000, 27: 409-422

[15] CramaY, SchynsM. Simulated annealing for complex portfolio selection problems [J]. European Journal of Operational Research, 2003, 150: 546-571

[16] DengX, LiZ, WangS. A minimax portfolio selection strategy with equilibrium [J]. European Journal of Operational Research, 2005, 166: 278-292

[17] HaoF F, LiuY K. Mean variance models for portfolio selection with fuzzy random returns [J]. Springer: J Appl Math Comput, 2007, 30: 9-38

[18] FineJ P, JiangH, ChappellR. On semi-competing risks data [J]. Biometrika, 2001, 88(4): 907-919

[19] CeriaS. To optimize or not to optimize: Is that the question? [M]//. The oxford handbook of quantitative asset management, 2012New YorkOxford University Press50-64

[20] SaxenaA, MartinC, StubbsR A. Aligning alpha and risk factors, a panacea to factor alignment problems? [R]. Axioma Inc., 2012

[21] GharakhaniM, SadjadiS. A new approach to robust modeling of the multi-period portfolio problem [J]. African Journal of Business Management, 2011, 5(23): 9998-10006

[22] SadjadiS J, GharakaniM, SafariE. Robust optimization framework for cardinality constrained portfolio problem [J]. Applied Soft Computing, 2012, 12(1): 91-99

[23] KolmP N, TüTüNcüR, FabozziF J. 60 years of portfolio optimization: Practical challenges and current trends [J]. European Journal of Operational Research, 2014, 234(2): 356-371

[24] SchaerfA. Local search techniques for constrained portfolio selection problems [J]. Computational Economics, 2002, 20(3): 177-190

[25] KellererH, MaringerD. Optimization of cardinality constrained portfolios with a hybrid local search algorithm [J]. OR Spectrum, 2003, 25(4): 481-495

[26] MaringerD. Portfolio management with heuristic optimization [J]. IEEE Computational Intelligence Magazine, 2008, 3(4): 31-34

[27] ChiamS C, TanK C, Al MamumA. Evolutionary multiobjective portfolio optimization in practical context [J]. International Journal of Automation and Computing, 2008, 5(1): 67-80

[28] ArmananzasR, LozanJ A. A multiobjective approach to the portfolio optimization problem [C]//. Proceedings of the IEEE Congress on Evolutionary Computation, 20051388-1395

[29] SoleimaniH, GolmakaniH, SalimiM H. Markowitz based portfolio selection with minimum transaction lots, cardinality constraints and regarding sector capitalization using genetic algorithm [J]. Expert Systems with Applications, 2009, 36: 5058-5063

[30] MarkowitzH. Portfolio Selection. Efficient diversification of Investments [M], 1959New YorkJohn Wiley & Sons

[31] CuraT. Particle swarm optimization approach to portfolio optimization [J]. Nonlinear Analysis: Real World Applications, 2009, 10: 2396-2406