Mutual funds trading strategy based on particle swarm optimization

被引：11

作者：

Hsu, Ling-Yuan ^{[1
]}

Horng, Shi-Jinn ^{[1
,2
,3
]}

He, Mingxing ^{[2
]}

Fan, Pingzhi ^{[3
]}

Kao, Tzong-Wann ^{[4
]}

Khan, Muhammad Khurram ^{[5
]}

Run, Ray-Shine ^{[6
]}

Lai, Jui-Lin ^{[6
]}

Chen, Rong-Jian ^{[6
]}

机构：

[1] Natl Taiwan Univ Sci & Technol, Dept Comp Sci & Informat Engn, Taipei 106, Taiwan

[2] Xihua Univ, Sch Math & Comp Engn, Chengdu 610039, Peoples R China

[3] SW Jiaotong Univ, Inst Mobile Commun, Chengdu 610031, Sichuan, Peoples R China

[4] Technol & Sci Inst No Taiwan, Dept Elect Engn, Taipei, Taiwan

[5] King Saud Univ, Ctr Excellence Informat Assurance, Riyadh 11451, Saudi Arabia

[6] Natl United Univ, Dept Elect Engn, Miaoli 36003, Taiwan

来源：

EXPERT SYSTEMS WITH APPLICATIONS | 2011年 / 38卷 / 06期

关键词：

Mutual funds; Moving average; Return on investment; Particle swarm optimization; ALGORITHM; CONVERGENCE;

D O I：

10.1016/j.eswa.2010.12.105

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Mutual funds have become the most popular products for diversity of investment, since they are able to disperse investment risks to the smallest degree. In selecting mutual funds, the past performance of funds plays a central role in the expectations of the future performance of funds. In 2008, the U.S. sub-prime broke out: numerous investors lost more than half of the capitals donated. Therefore, a good trading strategy is necessary. In this paper, a new funds trading strategy that combines turbulent particle swarm optimization (named TPSO) and mixed moving average techniques is presented and used to find the proper content of technical indicator parameters to achieve high profit and low risk on a mutual fund. The time interval of moving average of the proposed method is adjustable and the trading model could avoid and reduce loss by providing several good buy and sell points. We tested the proposed model using the historical prices of last 10 years and the experimental results show that the performance of the proposed model is far better than the best original performance. (C) 2010 Elsevier Ltd. All rights reserved.

引用

页码：7582 / 7602

页数：21

共 28 条

[1] Abraham A, 2009, STUD COMPUT INTELL, V203, P291
[2] ATIYA A, 1997, INT C NEUR NETW, V4, P2112
[3] Surveying stock market forecasting techniques - Part II: Soft computing methods
Atsalakis, George S.
Valavanis, Kimon P.
[J]. EXPERT SYSTEMS WITH APPLICATIONS, 2009, 36 (03) : 5932 - 5941
[4] Baba N., 1992, IJCNN International Joint Conference on Neural Networks (Cat. No.92CH3114-6), P371, DOI 10.1109/IJCNN.1992.287183
[5] The particle swarm - Explosion, stability, and convergence in a multidimensional complex space
Clerc, M
Kennedy, J
[J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 (01) : 58 - 73
[6] Investment using technical analysis and fuzzy logic
Dourra, H
Siy, P
[J]. FUZZY SETS AND SYSTEMS, 2002, 127 (02) : 221 - 240
[7] ELDER DA, 1993, TRADING LIVING
[8] Is smarter better? A comparison of adaptive, and simple moving average trading strategies
Ellis, Craig A.
Parbery, Simon A.
[J]. RESEARCH IN INTERNATIONAL BUSINESS AND FINANCE, 2005, 19 (03) : 399 - 411
[9] FORECASTING S-AND-P AND GOLD FUTURES PRICES - AN APPLICATION OF NEURAL NETWORKS
GRUDNITSKI, G
OSBURN, L
[J]. JOURNAL OF FUTURES MARKETS, 1993, 13 (06) : 631 - 643
[10] GUNASEKARAGE Abeyratna., 2001, Emerging Markets Review, V2, P17, DOI [https://doi.org/10.1016/S1566-0141(00)00017-0, DOI 10.1016/S1566-0141(00)00017-0]

← 1 2 3 →