用遗传-禁忌搜索混合算法求解组合投资问题  

A Hybrid Algorithm Based on Genetic Algorithm and Tabu Search to Solve Portfolio Problem

作　　者：王竹芳[1] 潘德惠[1] 

机构地区：[1]东北大学工商管理学院,辽宁沈阳110004

出　　处：《东北大学学报（自然科学版）》2006年第1期111-114,共4页Journal of Northeastern University(Natural Science)

基　　金：辽宁省自然科学基金资助项目(002012)

摘　　要：提出了一种基于遗传算法和禁忌搜索的混合算法,用遗传算法提供并行搜索的主框架,用禁忌搜索作为遗传算法的变异算子.遗传算法中变异过程解空间的搜索由禁忌搜索实现,并且用混合算法求解了概率准则意义下的组合证券投资模型.实例证明,遗传/禁忌混合算法有较强的爬山能力,较遗传算法有更高的计算效率,为组合证券投资者提供了一种高效的决策方法.A hybrid algorithm is proposed on the basis of genetic algorithm and tabu search, where the main flame of the algorithm is provided by genetic algorithm and the tabu search is taken as the mutation operator. Here the tabu search is used for the solution space in the process of mutation. The hybrid algorithm is devised to solve the portfolio investment model in terms of probability criterion. An example shows that the hybrid algorithm is superior to the early genetic algorithm in mountain-climbing ability and computing efficiency. So, the hybrid algorithm will provide an effective way in making decision to investors.

关 键 词：组合证券投资 遗传算法 禁忌搜索 混合算法 

分 类 号：F830[经济管理—金融学]