Dynamic Topology of Stock Correlation Networks from the Bull and Bear Perspective:a Case of Shanghai 50 Index

Chi XIE,Huidong BIAN,Gangjin WANG
DOI: https://doi.org/10.13306/j.1672-3813.2017.01.010
2017-01-01
Abstract:Daily data collected from Shanghai 50 Index constituent stocks from January 4, 2005 to December 31, 2008 is divided into three stages: bear I, bull and bear II.We study dynamic topology of stock correlation networks in each stage by using the minimal spannin tree (MST), hierarchical tree (HT) and main network property measures.The results show that: Industrial clustering exists and becomes more and more obvious in stock market;Manufacturing industry turns into the absolute center in bull market, which lasts until bear II market;Internal stocks of finance & insurance industry and steelmaking industry always maintain a high correlation, and the stocks of parent company and subsidiary and the stocks of the cross holdings companies are also close to each other;In addition, the main network property measures reveal that the structure of the stock market's correlation network is closer but worse in bull market than in bear markets.
What problem does this paper attempt to address?