我国股票关联网络的动态演化研究

发布时间：2018-07-06 20:07

本文选题：股票关联网络 + 最小生成树　；参考：《系统工程学报》2014年02期

【摘要】：利用最小生成树算法构建动态演化的我国股票关联网络.实证研究关联网络拓扑结构特征的动态演化规律、市场整体及个股价格行为与网络拓扑结构特征间的内在关系.结果表明:网络节点度和点强度服从无标度分布;随着时间的推进,股票间的价格波动关联关系呈现越来越强的稳定性;市场收益性越高及市场收益波动越小,网络平均距离就越大,股票间的价格波动关联性就越弱.较大的市场收益波动会导致网络变得更加收缩.绝大多数股票的收益率与其对应网络节点的度或点强度值呈负相关关系.市场中同行业的股票间易产生价格波动关联,验证了行业分类标准在投资组合构建中的参考价值.
[Abstract]:In this paper, the minimum spanning tree algorithm is used to construct the dynamically evolving stock association network in China. This paper empirically studies the dynamic evolution law of the topological structure characteristics of the related networks, and the intrinsic relationship between the price behavior of the market as a whole and the price behavior of individual shares and the topological characteristics of the networks. The results show that the network node degree and point strength clothing are scale-free distribution; with the advance of time, the relationship between the stock price volatility presents more and more stability, the higher the market profitability and the smaller the market return volatility, The greater the average distance of the network, the weaker the correlation between stock price fluctuations. Greater volatility in market returns can lead to more contraction of the network. The returns of most stocks are negatively correlated with the degree or point strength of the corresponding network nodes. It is easy to produce price fluctuation relation among the stocks of the same industry in the market, which verifies the reference value of the industry classification standard in the construction of the investment portfolio.
【作者单位】：东北大学工商管理学院;沈阳化工大学经济与管理学院;
【基金】：国家自然科学基金资助项目(71001022;71371044;71201108;71271047) 中国博士后科学基金资助项目(20100471460);中国博士后科学基金特别资助项目(2013T60295)
【分类号】：F832.51

【参考文献】