Scholarlink

<p>Exploring the application of deep learning in banking systemic risk prediction has important theoretical value and practical significance. For this purpose, we employ Long Short-Term Memory (LSTM) neural network model to predict the systemic risk contribution of banks in China. The results show that LSTM model has better prediction ability and generalization ability, and the prediction effect is significantly better than that of extreme gradient lifting tree, support vector machine and other reference models. Although LSTM model can achieve higher predictive accuracy than traditional models, it’s still a "black box" in terms of model interpretability. In order to open the "black box" of LSTM model, we use the SHapley Additive exPlanations (SHAP) method to explore the important factors affecting the systemic risk contribution.The results show that the main influencing factors of systemic risk contribution of different banks are different. FinTech and the weighted average interest rate of one-month inter-bank lending are important characteristic variables that affect the systemic risk contribution of banks.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Bank Systemic Risk Based on LSTM Model

  • 期刊论文

  • 发表日期：

    2025年5月15日

查看全文
  • Jiaxiang Huang,
  • Renxiang Wang

摘要

Exploring the application of deep learning in banking systemic risk prediction has important theoretical value and practical significance. For this purpose, we employ Long Short-Term Memory (LSTM) neural network model to predict the systemic risk contribution of banks in China. The results show that LSTM model has better prediction ability and generalization ability, and the prediction effect is significantly better than that of extreme gradient lifting tree, support vector machine and other reference models. Although LSTM model can achieve higher predictive accuracy than traditional models, it’s still a "black box" in terms of model interpretability. In order to open the "black box" of LSTM model, we use the SHapley Additive exPlanations (SHAP) method to explore the important factors affecting the systemic risk contribution.The results show that the main influencing factors of systemic risk contribution of different banks are different. FinTech and the weighted average interest rate of one-month inter-bank lending are important characteristic variables that affect the systemic risk contribution of banks.