The Building Of Financial Networks Based On Threshold And Its Applied Research

Financial market is a highly complex dynamics of complex systems, It helps to understand the internal structure of the financial system and the dynamic behavior with the correlation between different stocks. It needs to build a network before studying the structure and nature of the financial market. There are many methods to build the financial network, including the correlation coefficient method, most scholars considered to build a network using the traditional techniques to define a fixed threshold based on the correlation coefficient method. However, it found that the topology of the network is different along with the threshold selected. This paper first demonstrates the largest component’s status in financial markets and then put forward the concept of the effective threshold interval, and finally proposes concrete steps to build financial network with this method. It can be applied to practice.We study the construction of financial networks based on threshold and its application mainly from the following aspects.(1) With the analysis of the size of the largest component, we found it is a common phenomenon in financial market that the change of the largest component exits. It is the intrinsic properties of the financial network, with this, we determine the threshold based on the changes of the largest component in financial markets.(2) With the Pearson correlation coefficient distribution of three financial markets, we found it showed a certain degree of normality, which helps to determine the efficient range when building the financial network.(3) We give a method to build financial networks based on threshold.Finally, Shanghai financial network and Dow Jones financial network are built based on the method. It found that Shanghai financial network has motif and community structure obviously, and we analysis the possibility of investors to buy stocks based on motifs. Then we analysis the robustness and the investment risk of Dow Jones financial network under different thresholds.