Inter-bank Multilayer Network And Liquidity Risk Based On Big Data Of Media

In the report of the 19th National Congress of the Communist Party of China,General Secretary Xi Jinping emphasized that improving the financial supervision system and keeping the bottom line of systemic financial risk.In view of the core position of banks in the financial market,the stability of the banks is the key to the healthy and stable operation of the whole financial market.Moreover,because bank liquidity risk reflects the process of risk accumulation,credit risk,operational risk and market risk may become its inducements,which lead to more complex and higher regulatory requirements than other financial risks.In addition,media reports can not only provide big data of media for the study of liquidity risk in the banking system,but also interact with the related channels of liquidity network as the channel of influencing bank liquidity,and jointly affect the formation of inter-bank network and the contagion of bank liquidity risk.Therefore,based on the big data of media,it has important theoretical and practical reference value to deeply analyze the inter-bank multilayer network and influence mechanism of bank liquidity risk and its rescue strategy.Firstly,from the cross perspective of media sentiment and multilayer network,this paper constructs a model of inter-bank multilayer network including positive media sentiment network layer and negative media sentiment network layer.Then,this paper empirically analyzes the structural characteristics of inter-bank multilayer network of China,the importance of nodes in the inter-bank multilayer network and the existence of systemically important banks.The main conclusions of the empirical study are: the inter-layer correlation of multilayer network is high,but the inter-layer similarity of multilayer network is low.The inter-bank multilayer network has smaller average path length and larger clustering coefficient,showing the characteristics of small world network.Systemically important banks belong to large-scale state-owned banks and joint-stock commercial banks with large asset scale,but there are great differences in the order in different periods of time,especially some relatively small banks become systemically important banks in a specific period of time.In addition,in different media sentiment networks,the relationship between information centrality and bank asset size is not strictly positive.Secondly,based on the established model of inter-bank multilayer network,this paper takes the information centrality of different media sentiments as the index of network structure,and constructs different media sentiment indexes according to the obtained big data of media.Then,from the cross perspective of media sentiment and multilayer network,this paper analyzes the influence mechanism of different media sentiment,network structure and their interaction effects on bank liquidity risk respectively.The main conclusions of the empirical study are: the media sentiment index is significantly positively correlated with bank liquidity risk,but when only for national commercial banks,the negative media sentiment index and mixed media sentiment index have a negative impact on bank liquidity risk.Network information centrality has a significant negative correlation with bank liquidity risk,which reflects the dual effects of network on bank liquidity risk.Mixed interaction effect and negative interaction effect have significant negative impact on bank liquidity risk,while positive interaction effect has significant positive impact on bank liquidity risk.In addition,the impact of national commercial banks on bank liquidity risk is significantly higher than that of the whole bank,but not significantly stronger than that of other banks.Finally,from the cross perspective of media sentiment and multilayer network,this paper constructs endogenous inter-bank multilayer network and single-layer network,which are substantially consistent with the network structure of empirical research,and constructs a rescue strategy model of bank liquidity risk contagion including different rescue subjects.Then,we simulate and analyze the influence of rescue strategy on bank liquidity risk contagion in multilayer and single-layer networks.The main conclusions obtained from the simulation analysis are: in the inter-bank single-layer network,the random correlation method is more conducive to the contagion of bank liquidity risk,but the opposite is true in the inter-bank multilayer network.Compared with the inter-bank single-layer network,the contagion range of bank liquidity risk will be smaller in the inter-bank multilayer network.In addition,in the inter-bank single-layer network,by properly adjusting the factors of media sentiment,we can basically achieve the purpose of restraining the large-scale contagion of bank liquidity risk,but the implementation conditions are stricter,and the effect is more significant in the inter-bank multilayer network.In the inter-bank multilayer network,we can achieve the effect of quickly restraining the substantial contagion of bank liquidity risk by comprehensively adjusting the factors of media sentiment and rescue strategy.According to the saying all above,through the research of this article,can not only have a clearer understanding of the structural characteristics of inter-bank multilayer network of China,but also reflect the public’s sentiment tendency towards bank liquidity risk and the correlation degree of bank liquidity risk at different stages from the field of public wisdom.Then,this article provides theoretical and practical reference for preventing the bad impact of excessive market fanaticism and frequent investor speculation on bank liquidity risk,and promoting different rescue subjects to formulate more targeted and scientific rescue strategies.