基于双重期望效用的投资组合模型及其智能算法研究

发布时间：2018-01-03 12:41

本文关键词：基于双重期望效用的投资组合模型及其智能算法研究　出处：《宁夏大学》2017年硕士论文　论文类型：学位论文

【摘要】：一般的效用函数没有考虑历史事件在将来出现的可能性差异,即认为在历史中发生的概率也会以相同的概率在未来时间发生,为了避免这种很强的假设,本文研究了基于双重期望效用下的投资组合模型,并应用人工智能算法对其求解.因此,基于双重期望效用理论的效用函数来度量投资行为的效用思想下,本文的主要工作如下:1根据实际金融市场的需求,以证券的收益率服从正态分布作为前提,引入可投资数目最大上限的约束,建立相应的多目标投资组合优化模型,采用罚函数法将多目标投资组合模型转化为单目标模型,同时构造出符合该模型的教与学算法对其求解,并选取10支股票进行仿真实验.2引入限制性卖空的约束条件,建立了投资组合优化模型,同时引用权重系数作为风险厌恶因子,使得模型更加符合投资者的决策心理,从而确保了决策方案的可行性.此外,本文设计了适合模型的混沌鸟群算法求解模型,并与粒子群、教与学算法作了比较,得到了当风险厌恶因子A取不同值时混沌鸟群算法具有更好的优化效果,同时为决策者们提供了更佳的投资方案.3考虑到我国真实的证券市场存在一些摩擦因素,会受到最小交易量及交易费用的约束,同时站在投资者角度,必须以分散风险为一目标,设置了投资上限限制,使得本文所给出的模型更贴近金融市场中投资者们的决策行为,这一模型方案合理、可行,使投资者多一种选择方案;此外,设计了求解模型的粒子群算法,得到符合理论依据的数值结果,为投资者们提供一种最优选择.
[Abstract]:The general utility function does not take into account the possibility of historical events in the future, that is, the probability of occurrence in history will also occur in the future with the same probability, in order to avoid this strong hypothesis. In this paper, the portfolio model based on dual expected utility is studied and solved by artificial intelligence algorithm. Therefore, the utility function based on dual expected utility theory is used to measure the utility of investment behavior. The main work of this paper is as follows: 1. According to the demand of the actual financial market, taking the yield of securities as the premise of normal distribution, we introduce the constraint of the maximum upper limit of the number of investments that can be invested. The corresponding multi-objective portfolio optimization model is established and the penalty function method is used to transform the multi-objective portfolio model into a single-objective model. At the same time, a teaching and learning algorithm is constructed to solve the model. At the same time, 10 stocks are selected to carry on the simulation experiment. 2. The restrictive short selling constraint condition is introduced, and the portfolio optimization model is established. At the same time, the weight coefficient is used as the risk aversion factor. Make the model more in line with investors' decision-making psychology, so as to ensure the feasibility of the decision-making scheme. In addition, this paper designed a model suitable for the model of chaotic bird swarm algorithm to solve the model, and particle swarm optimization. Compared with the learning algorithm, the chaotic bird swarm algorithm has better optimization effect when the risk aversion factor A takes different values. At the same time, it provides policy makers with a better investment scheme .3 considering that there are some frictional factors in the real securities market in China, it will be constrained by the minimum transaction volume and transaction costs, and at the same time, it will stand in the perspective of investors. In order to make the model more close to the decision behavior of investors in the financial market, we must set the upper limit of investment with the goal of dispersing risk. This model is reasonable and feasible. One more option for investors; In addition, a particle swarm optimization algorithm is designed to solve the model, and the numerical results are obtained according to the theoretical basis, which provides an optimal choice for investors.
【学位授予单位】：宁夏大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F832.51;TP18

【参考文献】