A Network Model of Investment and Its Robustness Based on the Intrinsic Characteristics of Subjects in Stock Market
A Network Model of Investment and Its Robustness Based on the Intrinsic Characteristics of Subjects in Stock Market
复制标题
DOI：
--
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
Zhuang Ya-ming
中科院分区：
文献类型：
--
作者：
Zhuang Ya-ming
复杂网络分析已成为研究真实的世界中复杂系统共性的通用方法，并已被统计物理、社会科学和生物科学等领域所接受。复杂网络理论已被用于通过复杂网络理论来分析经济和金融系统的复杂性。基于投资者能力的内在特征，摘要构造了一个广义网络模型及其扩展，用于分析股票之间的价格收益和所有权相关性。第一部分介绍了基于节点属性的网络构造方法和网络的统计特性分析;第二部分讨论了投资网络的特征描述，并进一步提出了投资网络的构造算法和网络统计参数的理论分析;第三部分回顾了度分布、聚类系数、平均路径长度、仿真结果表明，两种网络都具有小世界和无标度特性，在可拓网络中，随着网络规模的递增，聚类系数不依赖于网络规模，平均路径呈现线性斜率。进一步分析了投资网络在随机攻击和选择性攻击两种情况下的鲁棒性，结果表明投资网络对随机攻击的鲁棒性要高于对选择性攻击的鲁棒性。总之，复杂网络分析是研究复杂系统复杂性的有效方法，在经济和金融领域将得到越来越多的应用，特别是对于研究不同系统的动力学机制有着重要的意义。
Complex network analysis has become a general method for studying common properties of complex systems in the real world,and has been accepted in the fields of statistical physics,social sciences,and biological sciences.Complex network theories have been adopted to analyze the complexity of economic and financial systems through complex network theory.Based on the intrinsic characteristics of investors' capacity,a generalized network model and its extension are constructed to analyze price earning and ownership correlations among stocks. The first part introduces the method of constructing a network based on nodal attributes and analyzing the statistical feature of a network.The second part discusses a brief description of investment network features.We further propose the algorithm of constructing an investment network and theoretically analyzing network statistical parameters.The third part review literatures related to statistical characteristics of degree distribution,clustering coefficient,average path length,and simulation analysis.The simulation results show that both networks exhibit small world and scale-free property.As the network size increases logarithmically in the extension network,clustering coefficient is not dependent on the network scale and the average path shows linear slope. We further analyze the robustness of the investing network under two conditions: random attack and selective attack.The results show that the investment network has a higher robustness for random attack than for selective attack. In summary,complex networkanalysis is an efficient way to study the complexity of complex systems and will be more and more utilized in the field of economy and finance.This analysis method is especially useful to study the dynamic mechanism of different systems.