研究带有基数限制的离散多因素投资组合模型.与传统的投资组合模型不同的是,该模型中投资组合的决策变量是交易手数(整数),且限制资产投资的最大数目,其最优化模型是一个非线性整数规划问题.分别用随机产生的一组数据和来自纳斯达克的40只股票数据,利用拉格朗日松弛的混合分枝定界算法求解此模型,并用FORTRAN语言编程,数值结果表明该算法能有效求解此模型.%The multi-factor portfolio selection model with cardinality constraints is considered in this paper.This discrete portfolio model is of integer quadratic programming problems.The branch-and-bound algorithm on the Lagrangian dual relaxation and continuous relaxation is exploited for this model.Computational experiments are carried out with data from Nasadaq stock market and randomly generated.The results show that the algorithm is capable of solving the problems with up to 80 securities.
展开▼
