Measuring the resilience of higher-order financial risk networks: A new interpretable multidimensional method

This paper proposes a flexible framework for measuring resilience in higher-order financial risk networks, emphasizing multidimensionality and interpretability. The framework comprises four modules: (i) constructing higher-order networks via the reconstruction of the higher-order structure of time series (RHOSTS) method; (ii) computing multidimensional topological metrics; (iii) estimating time-varying impulse responses of these metrics to exogenous shocks via a time-varying parameter vector autoregression (TVP-VAR); and (iv) deriving the corresponding measures of resistance (absorption intensity) and recovery (absorption duration). We apply the framework to examine the resilience of the global cross-border clean-energy risk network under ESGUI shocks. Empirical results indicate that network-level resilience is highly time-varying and that countries exhibit pronounced heterogeneity in their resilience. Different topological metrics capture distinct facets of resilience, thus mitigating the limitations of relying on any single indicator. By substituting alternative topological metrics and shock variables, the framework can flexibly adapt to a wide range of higher-order financial risk networks, offering strong generalizability and scalability.