This paper measures China's financial resilience from the perspective of external risk shocks and analyzes its influencing factors for forecasting. First, we introduce an innovative financial resilience model comprising three submodels: the dynamic factor model, the TVP-VAR model, and a resilience characteristic measurement model that captures resistance and recoverability through absorption intensity and absorption duration. The results show a clear inverse relationship between absorption intensity and absorption duration, with resilience fluctuations exhibiting distinct phase characteristics. Notably, intervals of low resilience often correspond to specific risk events. Second, we apply the Lasso-logistic model for recursive estimation and forecasting financial resilience, while comparing its performance to that of the Logistic regression model. The results indicate that the Lasso-logistic model achieves, on average, a 10 % higher forecasting accuracy than the Logistic model does. Among the most important features identified by the model are macroeconomic and public expectation variables. The analysis shows that the stability of economic fundamentals and market participants' confidence in the future play pivotal roles in strengthening financial resilience and ensuring the stability of the financial system.