具有增益的概率网络金融计划模型

本文绘出一类具有增益的概率网络金融计划模型．许多多阶段金融计划问题可纳入这类模型．在这类模型中，随机变量的分布函数与Alexander过滤交易规则密切联系在一起，金融市场交易信号由神经网络产生，目标函数的最优值按其期望值计算．文中提出临界流和临界路的概念，给出目标函数下界等于其期望值的充分必要条件和期望最优解的求解方法．

Probabilistic Network with Gain Model for Financial Planning

In this paper, a probabilistic network with gain model for financial planning is proposed and some multi-period planning problems in Finance are shown to be formulated in this model. In the model,the distributions of random variables are closely related to Alexander's filter trading rule and have practical implication in applying neural network to generate trading signals in a financial market. The optimum value of the objective function of the model is evaluated by its expected value. The concepts of critical flow and critical path are proposed, the necessary and sufficient conditions for the lower bound that gives the expected value are given, and the method locating the expected solution is provided.