Research on Influencing Factors of corporate financial risk based on LightBM-SHAP model: Differentiation analysis under different corporate nature

Renguang Cao; Jing Cao; Ting Huang

doi:10.62051/pfeb9x20

Authors

Renguang Cao
Jing Cao
Ting Huang

DOI:

https://doi.org/10.62051/pfeb9x20

Keywords:

financial risk; firm nature; feature selection; causal inference.

Abstract

With the deepening of economic globalization and the increasing complexity of enterprise operating environment, it is of great theoretical and practical significance to study the key factors affecting enterprise financial risk and their mechanisms for improving enterprise risk management level and market competitiveness. Based on the annual data of China's A-share listed companies from 2019 to 2023, this paper constructs A comprehensive evaluation index system of financial risk, and adopts entropy method to conduct a quantitative evaluation of corporate financial risk. Through the introduction of LightGBM model and Shapley value method, the paper conducts feature screening on the multi-dimensional factors that affect the financial risk of enterprises, and identifies the core indicators that are highly correlated with the financial risk. Then, from the Angle of enterprise nature, the paper discusses the mechanism of different influence of these indicators on financial risk in different enterprise types. The empirical results show that both financial and non-financial indicators have significant effects on the financial risk of enterprises, but the importance of these indicators is significantly different in different natures of enterprises, especially the influence of non-financial indicators in non-state-owned enterprises is more prominent. Based on the above research conclusions, this paper puts forward the corresponding policy suggestions, which provide a guiding reference for the financial risk management of enterprises.

Downloads

Download data is not yet available.

References

[1] Gu lisi. Research on financial risk early warning of listed companies in China's pharmaceutical manufacturing industry--Based on Stacking integration algorithm [J]. Small and Small and Medium-sized Enterprises Management and Technology,2023, (08): 164 - 166.

[2] WANG Chon, LIU Yali. A study on financial risk early warning by Benford and XGBoost model fusion - based on the data analysis of A-share listed companies in Shanghai and Shenzhen from 2000 to 2021 [J]. Research on Finance and Accounting, 2024, (01): 35 - 46.

[3] LI Xinhui, DONG Siping, XIAO Jingjing, et al. Research on financial risk assessment model of public hospitals based on factor analysis [J]. China Health Economics, 2010, 29 (10): 88 - 90.

[4] CAI Xinglin, Zhang Yaoga. Research on financial risk assessment and early warning of listed Chinese sporting goods companies: Based on financial data from 2009 to 2016 [J]. Journal of Xi 'a University of Physical Education, 2019, 36 (03): 282 - 289.

[5] WANG Zhuquan, SONG Xiaobin, WANG Yuanzhu. Evaluation and judgement of short-term financial risk of China's real economy--Comprehensive assessment and early warning of short-term financial risk considering stock and flow [J]. Management World, 2020, 36 (10): 156 - 170+216 - 222.

[6] Shen Qingyuan. Research on financial risk early warning management of automobile industry based on fuzzy comprehensive evaluation method--Taking JAC as an example [J]. Modern Business,2019, (35): 53 - 55.

[7] JANG Fuwei, LIN Yihao, MA Tian. Corporate bond default risk in the context of ‘de-rigging’: an investigation of machine learning early warning and economic mechanisms [J]. Financial Research,2023, (10): 85 - 103.

[8] Gepp A, Linnenluecke M K, O'Neill T, et al. Big Data Techniques in Auditing Research and Practice: Current Trends and Future Opportunities[J]. Social Science Electronic Publishing, 2017, 40: 102 - 115.

[9] ZHAO Qi, Xu Weijun, Ji Yucheng, Liu Guifang, Zhang Weiguo. Review on the application of Machine learning in financial asset price prediction and allocation [J]. Journal of Management, 20, 17 (11): 1716 - 1728.

[10] Mullainathan S, Spiess J. Machine Learning: An Applied Econometric Approach [J]. Journal of Economic Perspectives, 2017, 31(2): 87 - 106.

[11] Wang Fang, Wang Xuanyi, Chen Shuo. Machine Learning in economics research: Review and Prospect [J]. Journal of Quantitative and Technical Economics, 2019, 37 (04): 146 - 164.

[12] Ke Jian, Wang Qiqi. Research on early warning of financial risk in R&D-based manufacturing industry based on machine learning [J]. Management Accounting Research, 2023, (05): 22 - 32.

[13] Xingli Z, Wenjie W, Guochao L, et al. Optimizing Financial Risk Models in Digital Transformation-Deep Learning for Enterprise Management Decision Systems [J]. Journal of Organizational and End User Computing (JOEUC), 2024, 36 (1): 1 - 19.

[14] Guo Yaxin. Research on comprehensive evaluation of financial risk of small and medium-sized enterprises--Based on fuzzy comprehensive evaluation method [J]. Times Economy and Trade, 2021, 18 (09): 68 - 71.

[15] Gu X, Wang Y, Xing X, et al. The impact of green credit policy on heavily polluting enterprises’ financial risk: evidence from China [J]. Clean Technologies and Environmental Policy, 2024, 1 - 19.

[16] Wang, T., & Liu, X. (2019). "The Effects of Non-Financial Performance Indicators on Financial Risk: Evidence from China’s Technology Sector." Technological Forecasting and Social Change, 146, 620 - 628.

[17] Li, H., & Chen, Q. (2020). "Corporate Governance and Financial Risk: The Role of Non-Financial Factors." Corporate Governance: An International Review, 28 (6), 467 - 485.

[18] Chen, K., & Jiang, Y. (2021). "Financial Indicators and Their Impact on Financial Risk in Different Ownership Structures: Evidence from China." Emerging Markets Finance and Trade, 57 (8), 2255 - 2268.

[19] Altman, E, 1968, Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy0, Journal of Finance 23 (Sept.), pp. 589 - 609.

[20] WANG Shan, et al. Design and research on comprehensive evaluation model of water quality based on entropy value method [J]. Automation Application, 2024, 65 (12): 242 - 243+246.

[21] Liu Yunjing, Wu Bin, Zhang Min. Can independent directors identify the risk of corporate financial fraud - a study of predicting financial fraud based on machine learning [J]. Accounting Research, 2023, (09): 21 - 34.

[22] ZHANG Qian, MEI Yali, WANG Kui. Does ESG Performance Affect Corporate Financial Risk - Empirical Evidence Based on A-share Listed Companies [J]. Friends of Accounting, 2023, (21): 105 - 114.

[23] ZHOU Chen, ZHANG Feiteng, HAN Chen. Research on the impact of customer relationship on financial risk of listed companies [J]. Friends of Accounting, 2024, (07): 68 - 75.