Portfolio optimization considering cardinality constraints and based on various risk factors using the differential evolution algorithm

محورهای موضوعی : Financial Engineering

Behnaz Ghadimi ¹ , Mehrzad Minooei ² , Gholamreza Zomorodian ³ , Mirfeiz Fallahshams ⁴

1 - Department of Financial Management, Central Tehran Branch, Islamic Azad University, Tehran, Iran.
2 - Department of Industrial Management, Central Tehran Branch, Islamic Azad University, Tehran, Iran.
3 - Department of Finance, Central Tehran Branch, Islamic Azad University, Tehran, Iran.
4 - Department of Finance, Central Tehran Branch, Islamic Azad University, Tehran, Iran.

تاریخ دریافت : 1400/11/06 تاریخ پذیرش : 1401/04/30 تاریخ انتشار : 1402/12/07

کلید واژه: Cardinality constraint, Differential evolution algorithm, Value-at-risk, Conditional Value-at-Risk, Portfolio optimization,

چکیده مقاله :

As the main achievement of the modern portfolio theory, portfolio diversifica-tion based on risk and return has attracted the attention of many researchers. The Markowitz mean-variance problem is a convex quadratic problem turned into a mixed-integer quadratic programming problem when incorporating car-dinality constraints. Due to the high number of stocks in a market, this problem becomes an NP-hard problem. In this paper, a metaheuristic approach is pro-posed to solve the portfolio optimization problem with cardinality constraints using the differential evolution algorithm, while it is also intended to improve the solutions generated by the algorithm developed. In addition, variance, val-ue-at-risk, and conditional value-at-risk are assessed as risk measures. Candi-date models are solved for 50 top stocks introduced by the Tehran Stock Ex-change by considering the cardinality constraints of not more than five stocks within the portfolio and 24 trading periods. Finally, the obtained results are compared with the results of genetic algorithm. The results show that the pro-posed method has reached the optimal solution in a shorter time.

چکیده انگلیسی:

As the main achievement of the modern portfolio theory, portfolio diversifica-tion based on risk and return has attracted the attention of many researchers. The Markowitz mean-variance problem is a convex quadratic problem turned into a mixed-integer quadratic programming problem when incorporating car-dinality constraints. Due to the high number of stocks in a market, this problem becomes an NP-hard problem. In this paper, a metaheuristic approach is pro-posed to solve the portfolio optimization problem with cardinality constraints using the differential evolution algorithm, while it is also intended to improve the solutions generated by the algorithm developed. In addition, variance, val-ue-at-risk, and conditional value-at-risk are assessed as risk measures. Candi-date models are solved for 50 top stocks introduced by the Tehran Stock Ex-change by considering the cardinality constraints of not more than five stocks within the portfolio and 24 trading periods. Finally, the obtained results are compared with the results of genetic algorithm. The results show that the pro-posed method has reached the optimal solution in a shorter time.

منابع و مأخذ:


[1] Markowitz, H., Portfolio selection, The journal of finance, 1952; 7(1):77-91.doi: 10.1111/j.1540-6261.1952.tb01525.x

[2] Tehrani, R., Fallah Tafti, S., Asefi, S., Portfolio optimization using krill herd metaheuristic algorithm considering different measures of risk in tehran stock exchange, Financial research journal, 2018; 20(4):409-426. doi: 10.22059/FRJ.2019.244004.1006538
[3]Samuelson, P.A., The fundamental approximation theorem of portfolio analysis in terms of means, variances and higher moments, in Stochastic optimization models in finance. 1975; Elsevier: 215-220. doi: 10.1016/B978-0-12-780850-5.50023-X

[4]Markowitz, H., Portfolio selection: efficient diversification of investments, Cowies Foundation Monograph, 1959;16.
[5] Konno, H., Yamazaki, H., Mean-absolute deviation portfolio optimization model and its applications to Tokyo stock market, Management science, 1991; 37(5):519-531. doi: 10.1287/mnsc.37.5.519
[6] Young, M.R., A minimax portfolio selection rule with linear programming solution, Management science, 1998; 44(5): 673-683. doi: 10.1287/mnsc.44.5.673
[7] Jorion, P., Value at risk: the new benchmark for controlling market risk, 1997; Irwin Professional Pub.
[8] Rockafellar. R.T., Uryasev, S., Optimization of conditional value-at-risk, Journal of risk, 2000; 2: 21-42.
[9] Bienstock, D., Computational study of a family of mixed-integer quadratic programming problems, in International Conference on Integer Programming and Combinatorial Optimization. 1995;Springer. doi: 10.1007/3-540-59408-6_43
[10] Zamani, M., et al., An Expert System For Stocks Price Forecasting And Portfolio Optimization Using Fuzzy Neural Network, Fuzzy Modeling And Genetic Algorithm. 2015.
[11] Simaan, Y., Estimation risk in portfolio selection: the mean variance model versus the mean absolute deviation model, Management science, 1997; 43(10):1437-1446.
[12] Jorion, P., Value at risk: the new benchmark for managing financial risk, 2, 2001; McGraw-Hill New York.
[13] Nti, I.K., Adekoya, A.F., and Weyori, B.A., A systematic review of fundamental and technical analysis of stock market predictions, Artificial Intelligence Review, 2020; 53(4): 3007-3057. doi: 10.1007/s10462-019-09754-z
[14] Marti, G., Nielsen, F., Bińkowski, M., Donnat, P., A review of two decades of correlations, hierarchies, networks and clustering in financial markets, Progress in Information Geometry, 2021; 245-274. doi: 10.48550/arXiv.1703.00485
[15] Lukovac, V., Pamučar, D., Popović M.,, Đorović, B.,Portfolio model for analyzing human resources: An approach based on neuro-fuzzy modeling and the simulated annealing algorithm, Expert Systems with Applications, 2017; 90: 318-331. doi: 10.1016/j.eswa.2017.08.034
[16] Schaerf, A., Local search techniques for constrained portfolio selection problems, Computational Economics, 2002; 20(3): 177-190. doi: 10.1023/A:1020920706534
[17] Baykasoğlu, A., Yunusoglu, M.G., and Özsoydan F.B., A GRASP based solution approach to solve cardinality constrained portfolio optimization problems, Computers & Industrial Engineering, 2015; 90: 339-351. doi: 10.1016/j.cie.2015.10.009
[18] Chang, T.-J., Meade, N., Beasley, J.E., Sharaiha, Y.M., Heuristics for cardinality constrained portfolio optimisation, Computers & Operations Research, 2000; 27(13): 1271-1302. doi: 10.1016/S0305-0548(99)00074-X
[19] Ehrgott, M., Klamroth, K., Schwehm, C., An MCDM approach to portfolio optimization, European Journal of Operational Research, 2004; 155(3):752-770. doi: 10.1016/S0377-2217(02)00881-0
[20] Lin, C.-C. and Y.-T. Liu, Genetic algorithms for portfolio selection problems with minimum transaction lots, European Journal of Operational Research, 2008; 185(1): 393-404. doi: 10.1016/j.ejor.2006.12.024
[21] Mishra, S.K., Panda, G., and Majhi, R., Constrained portfolio asset selection using multiobjective bacteria foraging optimization, Operational Research, 2014; 14(1): 113-145. doi: 10.1007/s12351-013-0138-1
[22] Yin, X., Ni, Q., Zhai. Y., A novel PSO for portfolio optimization based on heterogeneous multiple population strategy, in 2015 IEEE Congress on Evolutionary Computation (CEC), 2015; IEEE. doi: 10.1109/CEC.2015.7257025
[23] Macedo, L.L., Godinho, P., Alves, M.J., Mean-semivariance portfolio optimization with multiobjective evolutionary algorithms and technical analysis rules, Expert Systems with Applications, 2017; 79: 33-43. doi: 10.1016/j.eswa.2017.02.033
[24] Kalayci, C.B., Ertenlice, O., Akyer, H., Aygoren, H., An artificial bee colony algorithm with feasibility enforcement and infeasibility toleration procedures for cardinality constrained portfolio optimization, Expert Systems with Applications, 2017, 85, P. 61-75. Doi: 10.1016/j.eswa.2017.05.018
[25] Kalayci, C.B., Polat, O., Akbay, M.A., An efficient hybrid metaheuristic algorithm for cardinality constrained portfolio optimization. Swarm and Evolutionary Computation, 2020, 54, P.100662.
[26] Yuen, M.-C., , Ng, S-C. Leung, Che, M.F., H., Metaheuristics for Index-Tracking with Cardinality Constraints, in 2021 11th International Conference on Information Science and Technology (ICIST). 2021. IEEE. doi: 10.1109/ICIST52614.2021.9440584
[27] Asoroosh, A., Atrchi, R., Ramtinnia, S., Portfolio Optimization Using Teaching-Learning Based Optimization (TLBO) Algorithm in Tehran Stock Exchange (TSE), Financial Research Journal, 2017; 19(2): 263-280. doi: 10.22059/JFR.2017.234738.1006462
[28] Cornuejols, G., Tütüncü, R., Optimization methods in finance, Cambridge University Press. 2006; 5.
[29] Storn, R., Price, K., Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces, Journal of global optimization, 1997; 11(4): 341-359. doi: 10.1023/A:1008202821328
[30] Dash, R., Rautray, R., Dash, R., Utility of a Shuffled Differential Evolution algorithm in designing of a Pi-Sigma Neural Network based predictor model, Applied Computing and Informatics, 2020. doi: 10.1016/j.aci.2019.04.001/full/html
[31] Deng, W., Shang, S., Cai, X., Zhao, H., Song, Y., Xu, J., An improved differential evolution algorithm and its application in optimization problem, Soft Computing, 2021; 25(7): 5277-5298. doi: 10.1007/s00500-020-05527-x
[32] Conover, W.J., Practical nonparametric statistics, john wiley & sons. 1999; 350.