房地产开发项目投资组合优化的改进蚁群算法

摘要/Abstract

摘要： 开发项目投资组合是分散房地产开发非系统风险的一重要策略。现有的房地产开发投资组合理论及算法存在缺陷,本文利用熵作为风险衡量指标,并将蚁群算法引入房地产开发领域,且针对基本蚁群算法存在的计算复杂,易陷入局部最优等缺陷,提出了一种基于信息熵的改进蚁群算法,采用由信息熵控制的路径选择及随机扰动策略实现了算法的自适应调节,克服了基本蚁群算法的不足。TSP问题的计算结果表明了该方法较之其他改进算法的优势。以各项目间的均值熵代替TSP中的各城市距离后的房地产投资组合计算实例表明,该方法具有较好的收敛性、稳定性和鲁棒性,是求解组合优化问题的一种较好的方法。

关键词: 房地产, 投资组合, 蚁群算法, 信息熵, 改进

Abstract: Portfolio is an import measure of risk distribution in real estate development.The theory and the algo-rithm of real estate have their shortcoming.We used information entropy to measure the risk.According to the deficiency of the basic Ant Colony Algorithm,specially it is easy to fall into the local extremum and its calculation is complicated.We proposed an improved Ant Colony Algorithm based on the information entropy.The informa-tion entropy is used to control the path selection and evolutional strategy by self-adjusting.Then the improved algorithm is introduced into the real estate portfolio.Application results have shown its efficiency and robustness in solving the real estate portfolio optimization problems.

Key words: real estate, portfolio, Ant Colony Algorithm, information entropy, improved

中图分类号:

F830

周书敬, 李彦苍. 房地产开发项目投资组合优化的改进蚁群算法[J]. 中国管理科学, 2004, (5): 74-79.

ZHOU Shu-jing, LI Yan-cang. Adaptive Ant Colony Algorithm Based on Information Entropy for Real Estate Portfolio Optimization[J]. Chinese Journal of Management Science, 2004, (5): 74-79.

参考文献

[1] James Vercammen. Irreversible investment under uncertainty and the threat of bankruptcy[J].Economics letters,2000, 66:319 - 325.
[2] 李华,等.熵一证券投资组合风险的一种新的度量方法[J].数学的实践与认识,2003,(6):16-21.
[3] Dorigo M, Maniezzo V, Colorni A. Ant system: optimization by a colony of coorperating agents [J]. IEEE Trans on SMC,1996,26(1) :28 - 41.
[4] Dorigo M,Gambardella L M.Ant colony system:a cooperative learning approach to the traveling salesman problem[J]. IEEE Trans on Evolutionary Computing, 1997, 1 ( 1 ): 53- 56.
[5] Colorni A,Dorigo M, Maniezzo V.Ant colony system for job - shop scheduling[J]. Belgian J. of Operations Research Statistics and Computer Science,1994,34(1 ) :39 - 53.
[6] 王笑蓉,吴铁军.Flowshop问题的蚁群优化调度方法[J].系统工程理论与实践,2003,7(5):65-71.
[7] Maniezzo V, Carbonaro A.An ANTS heuristic for the frequency assignment problem[J]. Future Generation Computer Systems,2000,16: 927 - 935.
[8] 张纪会,徐心和.一种新的模拟进化算法--蚁群算法[J].系统工程理论与实践,1999,(3):84-87.
[9] 许国志,顾基发,车宏安.系统科学[M].上海教育出版社,2000.
[10] 邱莞华.管理决策与应用熵学[M].机械工业出版社,2002.
[11] Dorigo M,Luca M.A study of Ant- Q[J].Proceedings of 4th International Conference on Parallel Problem from Nature,Berlin: Springer Verlag, 1996. 656 - 665.
[12] Stutzle T, Hoos H.MAX- min ant system[J].Future Generation Computer systems,2000,16:889 - 914.
[13] Lee SG, Jung TU, Chung TC.An effective dynamic weighted rule for ant colony system optimization[C]. Proc.of the 2001 Congress on Evolutionary Computation, IEEE Press, 2001.
[14] Tsai CF, Tsai CW.A new approach for soloving large salesman problem using evolution ant rules[C]. Neural Networks, IJCNN2002, Proc. Of the 2002 Int' l Joint Conf. on Honolulu,2002,2:1540 - 1545.
[15] Gambardella L M, Dorigo M. HAS - SOP: Hybrid ant system for the sequential problem[R]. Technical Report IDSIA, 1997.
[16] Kawamura, H. et al Improved multiple ant colonies system for Traveling Saleman Problems [J]. In Kozan, e. ohuchi,A. (eds.):Operations Research/Management Science at Work. Kluwer, Boston 2002,41 - 59.
[17] Kawamura, H. et al. Wultiple ant colony algorithm based on colony level interactions [J]. IEICE Transactions, Fundeamentals E83 - A 2000,371 - 379.
[18] LiYanjun, Wu Tiejun An adaptive ant colony system algorithm for continuous - space optimization problems [J].Journal of Zhejiang University SCIENCE, 2004,1: 40 -46.
[19] 陈峻,秦玲.具有感觉和知觉特征的蚁群算地[J].系统仿真学报,2003,15(10):1418-1425.
[20] 陈峻,沈洁,秦玲.蚁群算法进行连续参数优化的新途径[J].系统工程理论与实践,2003,(3):48-53.
[21] 吴庆洪,张纪会,徐心和.具有变异特征的蚁群算法[J].计算机研究与发展,1999,36(10):1240-1245.
[22] 曹浪财,罗键,李天成.智能蚁群算法--蚁群算法的改进[J].计算机研究,2003,10:62-64.
[23] 朱庆保,杨志军.基于变异和动态信息素更新的蚁群优化算法[J].软件学报,2004,15(2):185-192.