我国证券投资基金羊群行为的实证及演化博弈研究
发布时间：2018-11-15 17:50
【摘要】：本文运用实证检验和博弈分析方法对中国证券投资基金的羊群行为进行了分析研究,包括以下两项内容：(1)对中国证券投资基金的羊群行为进行了实证检验；(2)基于演化博弈理论对证券投资基金的羊群行为进行了博弈分析,并进行了模拟仿真。 首先,分析了羊群行为的概念,对其作了全面梳理,并将近年来国内外学者对羊群行为理论、模型等方面的研究成果进行了系统回顾。 实证检验方面,本文利用修正后的LSV检测法,选择深证和上证市场2010—2012年的100家开放式基金交易数据作为样本,进行计算分析,结果得出我国证券投资基金存在严重的羊群行为,其羊群行为度高达23.92%,其中买入羊群行为度要比卖出羊群行为度要小。同时得到,羊群行为与股票流动规模呈负相关,且在某些热门行业羊群行为程度更加严重。 之后引入博弈理论框架,构造证券市场常见情形下的博弈模型,并进行了博弈均衡分析。然后以基金经理为使自己的业绩排名靠前所得到的报酬和名誉等个人利益最大化为前提,构建了基金经理的演化博弈模型,并设定相应参数进行仿真演练,最后对仿真结果进行分析,得出仿真参数中的基金经理自主决策的收益、分析成本系数以及记忆长度均能影响羊群行为的产生。 最后,结合本文研究成果,从管理制度、投资环境、考核体系、提高市场参与者水平等角度提出了抑制或缓解羊群行为的对策措施与建议。
[Abstract]:This paper uses the empirical test and game analysis method to analyze the herd behavior of China's securities investment funds, including the following two aspects: (1) the herd behavior of China's securities investment funds is empirically tested; (2) based on evolutionary game theory, the herd behavior of securities investment funds is analyzed and simulated. Firstly, the concept of herding behavior is analyzed, and the research results of herding behavior theory and model are reviewed systematically. In the empirical test, this paper uses the modified LSV test method, selects 100 open-end fund transaction data of Shenzhen Stock Exchange and Shanghai Stock Exchange market from 2010 to 2012 as the sample, carries on the calculation analysis. The results show that there is serious herding behavior in the securities investment funds in China, and the herding behavior degree is as high as 23.92, among which the buying herding behavior is smaller than the selling herding behavior. At the same time, there is a negative correlation between herding behavior and the scale of stock flow, and the degree of herding behavior is more serious in some popular industries. Then the game theory frame is introduced to construct the game model under the common situation of the stock market, and the game equilibrium analysis is carried out. Then, based on the premise that fund manager maximizes his personal benefits, such as reward and reputation, the evolutionary game model of fund manager is constructed, and the corresponding parameters are set up for simulation exercise. Finally, the simulation results are analyzed, and the income of the fund manager in the simulation parameters is obtained. The cost coefficient and memory length can affect the herd behavior. Finally, combined with the research results of this paper, this paper puts forward some countermeasures and suggestions to restrain or alleviate herding behavior from the aspects of management system, investment environment, assessment system and raising the level of market participants.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51
本文编号：2334014
[Abstract]:This paper uses the empirical test and game analysis method to analyze the herd behavior of China's securities investment funds, including the following two aspects: (1) the herd behavior of China's securities investment funds is empirically tested; (2) based on evolutionary game theory, the herd behavior of securities investment funds is analyzed and simulated. Firstly, the concept of herding behavior is analyzed, and the research results of herding behavior theory and model are reviewed systematically. In the empirical test, this paper uses the modified LSV test method, selects 100 open-end fund transaction data of Shenzhen Stock Exchange and Shanghai Stock Exchange market from 2010 to 2012 as the sample, carries on the calculation analysis. The results show that there is serious herding behavior in the securities investment funds in China, and the herding behavior degree is as high as 23.92, among which the buying herding behavior is smaller than the selling herding behavior. At the same time, there is a negative correlation between herding behavior and the scale of stock flow, and the degree of herding behavior is more serious in some popular industries. Then the game theory frame is introduced to construct the game model under the common situation of the stock market, and the game equilibrium analysis is carried out. Then, based on the premise that fund manager maximizes his personal benefits, such as reward and reputation, the evolutionary game model of fund manager is constructed, and the corresponding parameters are set up for simulation exercise. Finally, the simulation results are analyzed, and the income of the fund manager in the simulation parameters is obtained. The cost coefficient and memory length can affect the herd behavior. Finally, combined with the research results of this paper, this paper puts forward some countermeasures and suggestions to restrain or alleviate herding behavior from the aspects of management system, investment environment, assessment system and raising the level of market participants.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51
【参考文献】
相关期刊论文 前4条
1 张宗强,伍海华;基于上证180指数股票的羊群行为实证研究[J];财经理论与实践;2005年01期
2 唐莉;张宗成;;基于我国证券市场羊群行为的宏观与微观策略[J];当代经济科学;2006年03期
3 吴福龙,曾勇,唐小我;中国证券投资基金的羊群行为分析[J];管理工程学报;2004年03期
4 孟卫东;王永平;;我国证券投资者交易行为的进化博弈与多重均衡研究[J];管理工程学报;2006年04期
，本文编号：2334014
本文链接：https://www.wllwen.com/guanlilunwen/bankxd/2334014.html
