Measuring systemic risk contributions in the Chinese sector using the dynamic tail dependence with asymmetric long memory

In the post-epidemic era, as China's economy slows and the risk exposure of various sectors gradually increases, research on the risk contribution of sectors should be strengthened. This study introduces the TVM-MIDAS Copula-MES approach, which considers tail dependencies with asymmetric long memory when capturing the systemic risk contribution of Chinese sectors. We find that incorporating the asymmetric long memory feature of tail dependence can enhance the accuracy of both model fitting and measurement of systematic risk contribution. Furthermore, aside from the consumer sector, all other sectors show some degree of tail-dependent long memory. Compared to other sectors, real estate displays significant long memory in both upper and lower tail dependencies, while the materials and medical sectors exhibit the strongest long memory in upper and lower tail dependencies, respectively. Additionally, during the US subprime crisis, the Chinese financial sector contributed significantly more to systemic risk than other sectors, whereas during the Chinese stock market crash, its contribution was the opposite. Throughout the COVID-19 pandemic, the systemic risk contribution of all Chinese sectors remained relatively stable. Moreover, the Chinese utilities sector has the smallest systematic risk contribution overall, while the information technology sector shows the opposite trend.