Abstract
In this paper, we analyze the relationship among stock networks by focusing on the statistically reliable connectivity between financial time series, which accurately reflects the underlying pure stock structure. To do so, we firstly filter out the effect of market index on the correlations between paired stocks, and then take a t-test based P-threshold approach to lessening the complexity of the stock network based on the P values. We demonstrate the superiority of its performance in understanding network complexity by examining the Hong Kong stock market. By comparing with other filtering methods, we find that the P-threshold approach extracts purely and significantly correlated stock pairs, which reflect the well-defined hierarchical structure of the market. In analyzing the dynamic stock networks with fixed-size moving windows, our results show that three global financial crises, covered by the long-range time series, can be distinguishingly indicated from the network topological and evolutionary perspectives. In addition, we find that the assortativity coefficient can manifest the financial crises and therefore can serve as a good indicator of the financial market development.
Highlights
Small-world[1] and scale-free[2] properties are two universal features found in analyzing real-world complex networks
We look for the best available datasets and obtain the daily data of 1532 stocks included in the Main Board (Excluding Depositary Receipts and Investment Companies) of the Hong Kong stock market, from January 2000 to July 2015, with 4060 trading days in total
Due to the complexity of this giant network of 1,279 stocks trading on the Main Board of the Hong Kong market, we only extract a small part (containing 48 components of Hang Sang Index (HSI) from the Main Board) and exhibit it in Fig. 1 for visualization
Summary
Small-world[1] and scale-free[2] properties are two universal features found in analyzing real-world complex networks. Through testing and comparing with the existing filtering methods, one motivation of this paper is to propose a new test-based approach that can filter out insignificant correlations but keep only the significantly correlated stock pairs in order to construct a reliable stock network. This approach is designed to understand the network complexity by accounting for the distribution of the correlation coefficients and setting a unified threshold (other than the correlation value threshold) . The other motivation of this paper is to verify the proposed approach and compare it with existing methods in terms of both static and dynamic performances
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
