›› 2019, Vol. 31 ›› Issue (8): 35-48.

• 经济与金融管理 • 上一篇    下一篇

中国地方金融风险空间关联网络及区域传染效应:2009-2016

沈丽, 刘媛, 李文君   

  1. 山东财经大学金融学院, 济南 250014
  • 收稿日期:2018-05-31 出版日期:2019-08-28 发布日期:2019-09-11
  • 通讯作者: 刘媛(通讯作者),山东财经大学金融学院博士研究生
  • 作者简介:沈丽,山东财经大学金融学院教授,博士生导师,博士;李文君,山东财经大学金融学院副教授,硕士生导师,博士。

China's Regional Financial Risk Spatial Correlation Network and Regional Contagion Effect: 2009-2016

Shen Li, Liu Yuan, Li Wenjun   

  1. School of Finance, Shandong University of Finance and Economic, Jinan 250014
  • Received:2018-05-31 Online:2019-08-28 Published:2019-09-11

摘要:

基于地方金融风险压力指数构建了2009-2016年我国省际地方金融风险空间关联网络,利用社会网络分析方法研究了地方金融风险空间关联网络的总体关联性,进一步进行中心性分析和块模型分析,同时考察网络结构对地方金融风险水平的效应。研究发现:我国地方金融风险空间关联网络是典型的“无标度网络”,各省份关联关系数具有分布不均匀性,同时还具有“小世界现象”;样本考察期内,空间关联程度呈现波动中上升的态势,地方金融风险存在明显的空间关联和传染效应;对整个金融风险关联网络来说,关联程度的上升会促进金融风险的传染,提高风险传染的破坏水平和影响范围。而在对网络结构特征的效应进行分析时发现,对个体省份来说,度数中心度、接近中心度和中间中心度的提高会降低金融风险水平,因此,中心性排名较低的省份应加强与其他省份的关联,中心性排名较高的省份应更加注重金融风险的防范和管理,以防发生系统性金融风险。

关键词: 地方金融风险, 空间关联, 传染效应, 社会网络分析

Abstract:

Based on regional financial risk stress index, this paper constructs the inter-provincial regional financial risk spatial correlation network from 2009 to 2016, uses social network analysis to examine the overall connectedness of regional financial risk spatial correlation network, and further carries out central analysis and block model analysis, and the effect network structure on the level of regional financial risks. The results show that, China's regional financial risk spatial association network is a typical "scale-free network" and associations in each province are unevenly distributed and featured by "small-world characteristics"; the degree of spatial correlation shows a rising trend in fluctuations and there are obvious spatial correlations and contagion effects in regional financial risks; for the entire financial risk related network, the increase in the degree of correlation will promote the spread of financial risks, increase the level of damage and scope of risk contagion. In the analysis of the effects of network structure characteristics, it is found that, for individual provinces, the increase in the degrees centrality, closeness centrality and betweenness centrality will reduce the level of financial risk. Therefore, the lower-ranking provinces should strengthen the association with other provinces and the provinces with higher central rankings should pay more attention to the prevention and management of financial risks to prevent systemic financial risks.

Key words: regional financial risk, spatial correlation, contagion effect, social network analysis