Early Warning Research on Systemic Financial Risk Based on BP Neural Networks
Zhaoran Zhuang, Yabo Duan · 2024
Systemic financial risk is closely linked to national financial stability and economic security. The ultimate consequence of a systemic financial risk event is a systemic financial crisis, which can disrupt the entire financial market and even the broader economic system. Therefore, enhancing the monitoring and analytical capabilities for systemic financial risks is crucial for effective financial management. This paper employs macroeconomic and market indicators to construct a BP neural network early warning model for predicting systemic risk. First, it conducts a theoretical examination of systemic financial risks. Next, based on this theoretical framework, it develops a warning indicator system comprising 14 economic and financial indicators. Subsequently, principal component analysis is applied to optimize and refine the selection of systemic financial risk indicators, resulting in four representative common factors. Finally, an empirical analysis is conducted using the BP neural network early warning model. The model is trained with data from the first quarter of 2010 to the fourth quarter of 2022, while the first quarter of 2022 to the third quarter of 2023 serves as the validation set. The trained BP neural network model is then used to predict the validation set. The results indicate that the actual and predicted values of the validation set are generally consistent, with a standard error of 0.411. This suggests that the model can accurately predict systemic financial risks.