Optimizing Portfolio Selection Using Harmony Search
This paper proposes a harmony search (HS) algorithm for cardinality constrained portfolio optimization problems with the objective of investment return maximization and risk minimization. To solve the problem, this study develops a hybrid encoding that mixes the integer and real variables. The generation of solutions is divided into three phases - harmony memory consideration, pitch adjustment, and random generation to get the balance between exploration and exploitation. This study tests the performance of the proposed algorithmic scheme on three global stock market indexes provided by the OR-Library. The results of HS also compare with simulated annealing (SA), tabu search (TS), and variable neighborhood search (VNS) algorithms in the literature. HS outperforms all the competing algorithms among all three test data sets and demonstrates its potential on portfolio optimization.
harmony search portfolio selection pareto optimum multiobjective optimization
Yun-Chia Liang Angela H. L. Chen Yu-Ting Chang
Department of Industrial Engineering and Management, Yuan Ze University,Taiwan, China Department of Finance Nanya Institute of Technology, Taiwan, China
国际会议
The Institute Industrial Engineera Asian Conference 2011(2011年国际工业工程师协会亚洲会议)
上海
英文
444-451
2011-06-10（万方平台首次上网日期，不代表论文的发表时间）