基于双重网络的人工股票市场中信息传播风险因子研究

发布时间：2018-06-19 15:17

本文选题：信息传播 + 双重网络　；参考：《湖南大学》2014年硕士论文

【摘要】：随着信息时代的到来，信息传播的地位和作用日趋重要，深刻地影响着我国社会的各个领域。尤其，金融市场中信息的传播会导致市场波动，甚至造成金融危机。由于金融市场中作为信息传播者的投资者行为的不可测性和复杂性以及信息传播途径的复杂性、多样性等因素，信息传播过程必然是复杂系统演化的过程，导致使用传统的数理方法很难描述市场投资者的微观行为和信息传播过程。虽然已有部分学者运用计算实验方法对信息传播进行了研究，但是由于投资者行为不可测性和信息传播复杂性的限制，导致了现有对股票市场信息传播风险的研究还不充分。针对股票市场信息传播风险的形成机理问题，本文利用计算实验方法，在分析总结前人研究成果的基础上，结合现实股票市场信息传播的特点，构建了基于双重网络的信息传播实验模型，进而建立仿真实验金融平台，以实验的方式深入研究股票市场信息传播风险问题。本文通过把仿真实验的结果与真实市场进行对比分析后认为，股价收益率呈现出尖峰厚尾的特征，并且具有较明显的波动聚集性，，验证模型是有效的。进一步讨论了风险因子对市场的流动性和波动性的影响方式及大小。研究发现，市场信息发布频率越小，信息透明度越高，市场的流动性越强，波动性越弱，此时市场的风险最小；而信息传播网络、交易者反馈信息的概率和学习速度对市场流动性和波动性的影响则没有得到一致的结果。最后，根据得出的实验结果，并针对我国信息披露制度存在的问题，提出了政策建议。
[Abstract]:With the coming of the information age, the status and function of information dissemination is becoming more and more important, which deeply affects every field of our society. In particular, the spread of information in the financial market will lead to market volatility, or even financial crisis. Due to the unpredictability and complexity of investors' behavior as information communicators in financial markets, and the complexity and diversity of information transmission channels, the process of information dissemination is bound to be a process of complex system evolution. It is difficult to describe the microcosmic behavior of market investors and the process of information dissemination by using traditional mathematical methods. Although some scholars have used computational experiments to study information dissemination, due to the unpredictability of investor behavior and the complexity of information dissemination, the existing research on the risk of information dissemination in stock market is not sufficient. In view of the formation mechanism of information communication risk in stock market, this paper uses computational experiment method, based on the analysis and summary of previous research results, combined with the characteristics of information dissemination in the real stock market. The experimental model of information dissemination based on dual network is constructed, and then the financial platform of simulation experiment is established. The risk of information dissemination in stock market is studied in depth by means of experiments. By comparing the results of the simulation experiment with the real market, it is concluded that the stock price yield shows the characteristics of peak and thick tail, and has obvious volatility and aggregation, which proves that the model is effective. The influence of risk factors on market liquidity and volatility is discussed. The study found that the smaller the frequency of market information release, the higher the transparency of information, the stronger the liquidity of the market and the weaker the volatility. The effect of feedback probability and learning speed on market liquidity and volatility is not consistent. Finally, according to the experimental results, and in view of the problems of information disclosure system in China, the policy recommendations are put forward.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F224;F830.91

【参考文献】