文献类型：期刊文献

英文题名：Constructing financial network based on PMFG and threshold method

作者：Nie, Chun-Xiao[1];Song, Fu-Tie[1]

机构：[1]East China Univ Sci & Technol, Sch Business, Dept Finance, Shanghai 200237, Peoples R China

年份：2018

卷号：495

起止页码：104

外文期刊名：PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS

收录：;EI(收录号：20175304592980);WOS:【SCI-EXPANDED(收录号:WOS:000424958600010)】；

基金：This research was partially supported by the National Natural Science Foundation of China (Fund Number: 71371073) and Shanghai Pujiang Program of China (Fund Number: 13PJC025).

语种：英文

外文关键词：Planar maximally filtered graph; Threshold network; Community structure; Cluster algorithm

摘要：Based on planar maximally filtered graph (PMFG) and threshold method, we introduced a correlation-based network named PMFG-based threshold network (PTN). We studied the community structure of PTN and applied ISOMAP algorithm to represent PTN in low dimensional Euclidean space. The results show that the community corresponds well to the cluster in the Euclidean space. Further, we studied the dynamics of the community structure and constructed the normalized mutual information (NMI) matrix. Based on the real data in the market, we found that the volatility of the market can lead to dramatic changes in the community structure, and the structure is more stable during the financial crisis. (C) 2017 Elsevier B.V. All rights reserved.