基于沪深300指数的交易者个人进化机制的实证研究

发布时间：2018-03-26 00:27

本文选题：人工股票市场　切入点：行为金融学　出处：《天津大学》2012年硕士论文

【摘要】：传统经典金融理论在理性人的假设基础上形成了其核心内容有效市场假说。根据这一假说发展起来的各种金融理论,包括资产组合理论、资本资产定价模型、套利定价模型以及期权定价模型等一起构成了传统经典金融理论的基础。然而,随着金融市场的发展,大量金融异象涌现出来,如,股票长期投资的收益率溢价、股票价格的异常波动与股价泡沫、封闭式基金之谜、股价对市场信息的过度反应或反应不足等等。这些异象向传统经典金融理论发起了挑战。其中较为核心的争论焦点为理性人的假设太过严格,众多研究结果表明,市场参与者很难达到完全理性的程度,而表现出了显著的有限理性特征。本文以市场参与者异质、有限理性的假设为基础,通过神经网络模型模拟交易者的决策规则,建立遗传算法模型模拟交易者的学习进化过程,由下而上的观察交易者的学习进化行为对股票市场产生的影响。研究结果发现,不同学习速度下,股票市场的波动性表现出了不同的特征。并以沪深300指数代表市场,实证检验了现实股票市场中不同学习速度下股市的波动状况。回归检验结果表明,换手率代表的学习速度对股市波动性的影响是显著的。实证和模拟结果说明,交易者的学习速度对股票市场有显著影响。
[Abstract]:On the basis of rational people's hypothesis, the traditional classical financial theory has formed its core content, efficient market hypothesis. According to this hypothesis, various financial theories, including portfolio theory, capital asset pricing model, have been developed. The arbitrage pricing model and the option pricing model together form the basis of the classical financial theory. However, with the development of the financial market, a large number of financial anomalies emerge, such as the return premium of long-term stock investment. The unusual fluctuation of stock price and the bubble of stock price, the riddle of closed-end fund, Stock prices overreact or underreact to market information, etc. These anomalies challenge the traditional classical financial theory. The core of the argument is that the hypothesis of rational people is too strict, and many studies show that, It is very difficult for market participants to reach the level of complete rationality, and show obvious characteristics of limited rationality. Based on the hypothesis of heterogeneity and limited rationality of market participants, this paper establishes genetic algorithm model to simulate the learning and evolution process of traders through the neural network model to simulate the decision-making rules of traders. A bottom-up look at how traders' learning and evolutionary behavior affects the stock market. The results show that the volatility of the stock market shows different characteristics at different learning speeds. The CSI 300 index is used to represent the market. The results of regression test show that the learning speed represented by turnover rate has a significant effect on the volatility of stock market. Traders' learning speed has a significant impact on the stock market.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;F224

【相似文献】