【作者】 安智宇；

【Author】 An Zhiyu Nanjin University of Science and Technology, Jiangsu2, 10094

【机构】 南京理工大学经济管理学院；

【摘要】 考虑了浮动佣金政策下交易成本函数的变化问题,非线性交易成本通常出现在交易量很大或者交易量很少时.对于既定风险条件下考虑非线性交易成本且期望效益最大化的非线性规划模型,将约束条件和目标函数最优化分离开,采用反馈神经网络系统运动方程处理约束问题,并利用遗传算法求解目标函数最优。通过实例说明该算法对所建立问题的求解是有效的.更多还原

【Abstract】 Nonlinear transaction costs frequentle occur in practice when the trade volume is either very small or very large.A nonlinear programme is constructed under transaction costs .The algorithm congsists of a hybrid neural-genetic algorithm, formed by a recurrent Neural network, which solves the problem’s constraints, and a genetic algorithm for optimizing the objective function.An illustration is given to show that the algorithm is effective.更多还原