Document Type : Original Article

Authors

1 Assistant Professor, Allameh Tabatabaei University

2 Associate Professor, Islamic Azad University, Central Tehran Branch

3 Allameh Tabatabaei University

Abstract

Optimal Portfolio Selection is one of the most important issues in the field of financial research. In the present study, we try to compare four various different models, which optimize three-objective portfolios using “Postmodern Portfolio Optimization Methods”, and then to solve them. These modeling approaches take into account both multidimensional nature of the portfolio selection problem and requirements imposed by investors. Concretely different models optimize the expected return, the down side risk, skewness and kurtosis given portfolio, taking into account budget, bounds and cardinality constrains. The quantification of uncertainty of the future returns on a given portfolio is approximated by means of LR-fuzzy numbers, while the moments of its returns are evaluated using possibility theory. In order to analyze the efficient portfolio, which optimize three criteria simultaneously, we build a new NSGAII algorithm, and then find the best portfolio with most Sortio ratio from the gained Pareto frontier. Thus, in this paper we choose 153 different shares from different industries and find their daily return for ten years from April of 2006 till March of 2017 and then we calculate their monthly return, downside risk, skewness, kurtosis and all of their fuzzy moments. After designing the four models and specific algorithm, we solve all of the four models for ten times and after collection of a table of the answers, compare all of them with Treyner ratio. At last, we find that using fuzzy and possibistic theory make higher return and better utilized portfolios.

Keywords

1. D.Dubois, H.Prade, (1980), Fuzzy Sets and Systems Theory and Applications,
2. Demic Aca-press, NewYork.
3. Reben Saborido, Ana B.Ruiz, Jose D.Bermudez, Enriqueta Vercher, Applied soft
computing 39 (2016) 48-63
4. C.A.C. Coello, G.B. Lamont, D.A.V. Veldhuizen, (2001), Evolutionary Algorithms for
Solving Multi-Objective Problems, 2
nd ed. Springer,New York,US,2007.
5. K. Deb, (2001), Multi-objective Optimization Chichester, Using Evolutionary Algorithms,
Wiley, 2001.
6. X. Zou, Y.Chen, M.Liu, L.Kang, (2008), a new evolutionary algorithm for solving manyobjective optimization problems, IEEE Trans. Syst.Man Cybern. B 38 (5)
7. 1402–1412.
8. K. Deb, A. Pratap, S. Agarwal, T. Meyarivan, (2002), a fast and elitist multiobjective
genetic algorithm: NSGA-II, IEEE Trans. Evol. Comput. 6 (2) 182–197.
9. K. Liagkouras, K. Metaxiotis, (2015), Efficient portfolio construction with the use of
multiobjective evolutionary algorithms: Int.J. Best practices and performance metrics, Inf.
Technol. Decis. Mak.14 (3) 535–564.
10.Mehdi Khajezadeh Dezfuli, (2016), Operation reaserch for "Specialty Industrial
Management Course code 2164" Ph.D. (In Persian), Sharif Masters
11.T.J.Chang, N.Meade, J.E.Beasley, strainedY.M.Sharaiha, Heuristics for cardinality con –
portfolio optimization, Comput. Oper. Res. 27(13) (2000)1271–1302.
12.D.Maringer, H.Kellerer, (2003), Optimization a hybrid of cardinality constrained portfolios
with local search algorithm, ORSpectr. 25(4) 481–495
13.R.Moral-Escudero, R.Ruiz-Torrubiano, A.Suarez, (2006), Selection of optimalinvestportfolios with cardinality Computation, constraints, in: IEEE Congresson Evolutionary,
pp.2382–2388.
14.P.Skolpadungket, K.Dahal, (2007), multi-objective Harnpornchai, Portfolio optimization
using genetic algorithms, putation, in: IEEE Congress on Evolutionary Com-, pp. 516–523.
15.S.C.Chiam, K.C. Tan, A.A. Mamum, (2008), Evolutionary multi-objective portfolio
optimization in practical context, Int.J.Atom.Computer.05 (1), 67-80
16.K.P. Anagnostopoulos, G.Mamanis, (2008), objectives a portfolio optimization model with
three and discrete variables, Int.J.Autom.Comput. 05 (1) 67–80Comput. Oper.Res. 37 (7)
(2010) 1285–1297.
17.K.P.Anagnostopoulos, G.Mamanis, (2011), the mean-variance cardinality constrained
portfolio optimization problem: an experimental evaluation of five multiobjec-evolutionary
algorithms, Expert Syst. Appl. 38(11), 14208-14217
18.E.Zitzler, M.Laumanns, L.Thiele, (2001), lutionary SPEA2: improving the strength Pareto
evo-algorithm for multiobjective D.T.Tsahalis, optimization, in: K.C. Giannakoglou, J.
Periaux, K.D. Papailiou, for Design T. Fogarty (Eds.), Evolutionary Methods Optimisation
and Control International with Application to Industrial Problems, Center for Numerical
pp.Methods in Engineering (CMINE), ,95–100