Identification of systemically important financial institutions in a multiplex financial network: A multi-attribute decision-based approach

The identification of systemically important financial institutions (SIFIs) is vital to prevent and control systemical financial risk. Recently, network-based methods have been widely used to identify SIFIs by considering the complex relationship of financial institutions. However, most existing network-based methodologies typically use a single measure. In this study, we first identify SIFIs from three different perspectives, including the market value of financial institutions, the correlation between financial institutions, and the risk spillover between financial institutions. Then, we first employ the stock price volatility and investor sentiment data to form a multiplex financial network with four layers based on variance decompositions and correlation-based methods. Moreover, by incorporating features of the multiplex financial network and institutions’ market value, we identify SIFIs with technique for order preference by similarity to an ideal solution (TOPSIS). Furthermore, we conduct an empirical analysis of the Chinese financial market from 2016 to 2020. The results indicate that national commercial banks and Internet financial securities play an essential role in the Chinese financial system during the period, while insurances are not sensitive to price volatility and investors’ sentiment. Compared with existing methods, the proposed approach is valid to identify SIFIs from multiple perspectives, which can be generalized to other regional financial markets for identifying SIFIs.