Research on the Role of ESG Indicators on Financial Risk Early Warning in Financial Industry Based on Deep Learning Modeling

Yueqi Wang

doi:10.54691/xttk7f32

Authors

Yueqi Wang

DOI:

https://doi.org/10.54691/xttk7f32

Keywords:

ESG indicators; financial risk early warning; deep learning; LSTM model; financial sector.

Abstract

With the promotion of green financial policies and the implementation of the "dual carbon" goal, ESG indicators have gradually become an important tool for measuring corporate sustainability. As a pioneer in information disclosure, the relationship between ESG performance and financial risk in the financial industry has attracted attention. This study takes Chinese listed companies in the financial industry as a sample, constructs a risk early warning system covering financial and ESG indicators based on 2019-2024 data, and introduces Logistic regression, Random Forest and LSTM deep learning models for comparison. The results show that the predictive performance of the three models is improved after the inclusion of ESG factors, with the LSTM model performing the best on indicators such as AUC and F1. Further analysis shows that the governance dimension of ESG contributes most significantly to risk early warning, while the environmental and social dimensions contribute relatively weakly. Based on the above results, the study proposes to incorporate ESG indicators into the risk management framework, improve the quality of information disclosure, and strengthen the application of AI technology and talent cultivation, in order to improve the risk identification and prevention and control capabilities of financial institutions.

Downloads

Download data is not yet available.

References

[1] Xiao Hongjun, Ding Fengru. ESG development in China: review of the 14th Five-Year Plan and outlook for the 15th Five-Year Plan [J]. Reform, 2025, (04): 44-62.

[2] ZHANG Xiaojing,ZHANG Ming,FEI Zhaoqi,et al. China's financial development under triple pressure [J]. Financial Review, 2022, 14 (01): 1-22+121.

[3] Korzeb Z, Karkowska R, Matysek-Jędrych A, et al. How do ESG challenges affect default risk? An empirical analysis from the global banking sector perspective[J]. Studies in Economics and Finance, 2025, 42(1): 89-114.

[4] Atif M, Ali S. Environmental, social and governance disclosure and default risk[J]. Business Strategy and the Environment, 2021, 30(8): 3937-3959.

[5] Zhang Lin,Pan Jiaying. Research on credit risk early warning of corporate bonds incorporating ESG factors [J]. Financial Theory Exploration, 2021, (04): 51-65.

[6] An Zhanran,Wang Shuai,Zhu Tingjun. Does ESG performance promote firms' exports? --Micro evidence from Chinese listed companies [J]. Southern Economy, 2024, (06): 138-158.

[7] Liu Xuejuan,Chang Ruyue,Zhang Jingyi,et al. A study of corporate default risk early warning considering ESG performance [J]. Finance Theory and Practice, 2023, (04): 45-57.

[8] Zhong Wei,Dong Miao. Intrinsic association between corporate financial characteristics and corporate ESG scores - An exploration of identification prediction based on random forest approach [J]. Accounting Research, 2024, (07): 175-192.

[9] Pan Zeqing. Logistic regression early warning model of corporate debt default risk [J]. Shanghai Economic Research, 2018, (08): 73-83.

[10] Duan Yixue, Peng Xuemei, Fang Kuangnan. A study on financial early warning of Chinese property insurance companies - based on random forest financial diagnosis [J]. Insurance Research, 2024, (04): 20-33.

[11] Xu K, Cheng Y, Long S, et al. Advancing financial risk prediction through optimized LSTM model performance and comparative analysis[C]//2024 IEEE 2nd International Conference on Sensors, Electronics and Computer Engineering (ICSECE). IEEE, 2024: 1264-1270.

[12] Shiri F M, Perumal T, Mustapha N, et al. A comprehensive overview and comparative analysis on deep learning models: CNN, RNN, LSTM, GRU[J]. arXiv preprint arXiv:2305.17473, 2023.