AI Chat Paper
Note: Please note that the following content is generated by AMiner AI. SciOpen does not take any responsibility related to this content.
{{lang === 'zh_CN' ? '文章概述' : 'Summary'}}
{{lang === 'en_US' ? '中' : 'Eng'}}
Chat more with AI
Powered by AMiner. Clicking this button will take you away from SciOpen.
Home Journal of Social Computing Article
View PDF
Cite
EndNote(RIS) BibTeX
Collect
Submit Manuscript AI Chat Paper
Show Outline
Outline
Show full outline
Hide outline
Outline
Show full outline
Hide outline
Open Access

Predicting Credit Bond Default with Deep Learning: Evidence from China

Ning Zhang1( ), Wenhe Li1( ), Haoxiang Chen1, Binshu Jia2, Pei Deng1
School of Information, Central University of Finance and Economics, Beijing 100081, China
SDIC Taikang Trust Co., Ltd., Beijing 100034, China
Show Author Information

Abstract

China’s credit bond market has rapidly expanded in recent years. However, since 2014, the number of credit bond defaults has been increasing rapidly, posing enormous potential risks to the stability of the financial market. This study proposed a deep learning approach to predict credit bond defaults in the Chinese market. A convolutional neural network (CNN) was selected as the classification model and to reduce the extreme imbalance between defaulted and non-defaulted bonds, and a generative adversarial network (GAN) was used as the oversampling model. Based on 31 financial and 20 non-financial indicators, we collected Wind data on all credit bonds issued and matured or defaulted from 2014 to 2021. The experimental results showed that our GAN+CNN approach had superior predictive performance with an area under the curve (AUC) of 0.9157 and precision of 0.8871 compared to previous research and other commonly used classification models—including the logistic regression, support vector machine, and fully connected neural network models—and oversampling techniques—including the synthetic minority oversampling technique (SMOTE) and Borderline SMOTE model. For one-year predictions, indicators of solvency, capital structure, and fundamental properties of bonds are proved to be the most important indicators.

Keywords

convolutional neural networkpredictiongenerative adversarial networkcredit bond defaultimbalanced data processing

References

[1]

C. Ma, H. Sun, S. Wang, S. Lu, and J. Li, Bond default prediction based on deep learning and knowledge graph technology, IEEE Access, vol. 9, pp. 12750–12761, 2021.
DOI Google Scholar
[2]

T. Shumway, Forecasting bankruptcy more accurately: A simple hazard model, J. Bus., vol. 74, no. 1, pp. 101–124, 2001.
DOI Google Scholar
[3]

B. P. Foster, T. J. Ward, and J. Woodroof, An analysis of the usefulness of debt defaults and going concern opinions in bankruptcy risk assessment, J. Account. Auditing Finance, vol. 13, no. 3, pp. 351–371, 1998.
DOI Google Scholar
[4]

J. B. Warner, Bankruptcy costs: Some evidence, J. Finance, vol. 32, no. 2, pp. 337–347, 1977.
DOI Google Scholar
[5]

X. Yao, K. Li, and L. Yu, Corporate bond default risk prediction based on generative adversarial networks oversampling technique under unbalanced samples, (in Chinese), Systems Engineering Theory & Practice, no. 10, pp. 2617–2634, 2022.
Google Scholar
[6]

W. H. Beaver, Financial ratios as predictors of failure, J. Account. Res., vol. 4, pp. 71–111, 1966.
DOI Google Scholar
[7]

J. Wu, Y. Zhang, and X. Wang, Quantitative study on enterprise bond default risk under information disclosure distortion, Mathematical Statistics and Management, no. 1, pp. 175–190, 2017.
Google Scholar
[8]

E. I. Altman, Financial ratios, discriminant analysis and the prediction of corporate bankruptcy, J. Finance, vol. 23, no. 4, pp. 589–609, 1968.
DOI Google Scholar
[9]

X. Tang, S. Li, M. Tan, and W. Shi, Incorporating textual and management factors into financial distress prediction: A comparative study of machine learning methods, J. Forecast., vol. 39, no. 5, pp. 769–787, 2020.
DOI Google Scholar
[10]

D. Wang, Z. Wu, and B. Zhu, Controlling shareholder characteristics and corporate debt default risk: Evidence based on machine learning, Emerg. Mark. Finance Trade, vol. 58, no. 12, pp. 3324–3339, 2022.
DOI Google Scholar
[11]

G. Wang, J. Ma, G. Chen, and Y. Yang, Financial distress prediction: Regularized sparse-based random subspace with ER aggregation rule incorporating textual disclosures, Appl. Soft Comput., vol. 90, p. 106152, 2020.
DOI Google Scholar
[12]

Y. Huo, An empirical study on credit rating and credit default based on machine learning algorithms, (in Chinese), Financial Regulation Research, no. 3, pp. 15–35, 2022.
Google Scholar
[13]

S. Fallahpour, E. N. Lakvan, and M. H. Zadeh, Using an ensemble classifier based on sequential floating forward selection for financial distress prediction problem, J. Retail. Consum. Serv., vol. 34, pp. 159–167, 2017.
DOI Google Scholar
[14]

S. Khemakhem and Y. Boujelbene, Predicting credit risk on the basis of financial and non-financial variables and data mining, Rev. Account. Finance, vol. 17, no. 3, pp. 316–340, 2018.
DOI Google Scholar
[15]

Y. Wang, L. Zhou, and D. Zhang, Corporate debt default risk prediction: A machine learning perspective, (in Chinese), Fiscal Science, no. 6, pp. 62–74, 2022.
Google Scholar
[16]

H. Qian, B. Wang, M. Yuan, S. Gao, and Y. Song, Financial distress prediction using a corrected feature selection measure and gradient boosted decision tree, Expert Syst. Appl., vol. 190, p. 116202, 2022.
DOI Google Scholar
[17]

K. H. Kim and S. Y. Sohn, Hybrid neural network with cost-sensitive support vector machine for class-imbalanced multimodal data, Neural Netw., vol. 130, pp. 176–184, 2020.
DOI Google Scholar
[18]

R. Becerra-Vicario, D. Alaminos, E. Aranda, and M. A. Fernández-Gámez, Deep recurrent convolutional neural network for bankruptcy prediction: A case of the restaurant industry, Sustainability, vol. 12, no. 12, p. 5180, 2020.
DOI Google Scholar
[19]
X. Chen, J. Wu, and M. Xu, Measurement and prevention of individual default risk of credit debt in China based on LSTM deep learning model, (in Chinese), Fudan Journal (Social Sciences Edition), vol. 63, no. 3, pp. 159–173, 2021.
[20]

T. Hosaka, Bankruptcy prediction using imaged financial ratios and convolutional neural networks, Expert Syst. Appl., vol. 117, pp. 287–299, 2019.
DOI Google Scholar
[21]
Y. Tan, J. Chen, and J. Sun, Research on financial crisis warning of listed companies based on CNN, (in Chinese), Journal of Southwest China Normal University (Natural Science Edition), vol. 46, no. 5, pp. 73–80, 2021.
DOI
[22]

Y. Zhang, Research on early warning of credit default risk based on unbalanced samples, (in Chinese), South China Finance, no. 1, pp. 5–14, 2019.
Google Scholar
[23]
P. Soltanzadeh and M. Hashemzadeh, RCSMOTE: Range-controlled synthetic minority over-sampling technique for handling the class imbalance problem, Inf. Sci., vol. 542, pp. 92–111, 2021.
DOI
[24]

U. Fiore, A. De Santis, F. Perla, P. Zanetti, and F. Palmieri, Using generative adversarial networks for improving classification effectiveness in credit card fraud detection, Inf. Sci., vol. 479, pp. 448–455, 2019.
DOI Google Scholar
[25]

J. Zhang and S. Liu, Pay restraint, return on human capital investment and firm performance in state-owned enterprises, (in Chinese), Communication of Finance and Accounting, no. 14, pp. 38–41, 2020.
Google Scholar
[26]

M. Jiang, W. Zhou, J. Shi, and X. Li, Research on the influencing factors of bond default of listed enterprises: Empirical analysis based on fsQCA, (in Chinese), Chinese Journal of Management, vol. 18, no. 7, pp. 1076–1085, 2021.
Google Scholar
[27]

H. Kuang, H. Du, and H. Feng, Construction of the credit risk indicator evaluation system of small and medium-sized enterprises under supply chain finance, (in Chinese), Science Research Management, vol. 41, no. 4, pp. 209–219, 2020.
Google Scholar
[28]

Y. Xiao and Y. Xiang, Early warning of enterprise bond default risk: Based on GWO-XGBoost method, (in Chinese), Shanghai Finance, no. 10, pp. 44–54, 2021.
Google Scholar
[29]
Y. Wu, L. Zhai, and S. Wu, Hurting the innocent: The role of bond mutual funds in propagating idiosyncratic bond defaults, (in Chinese), Journal of Xiamen University (Arts & Social Sciences), no. 4, pp. 41–52, 2019.
[30]

X. Zhang and Y. Liu, Research on regional contagion effect of bond default: Based on bond secondary market data, (in Chinese), Finance Research, no. 3, pp. 22–31, 2021.
Google Scholar
[31]

S. Xu, and Y. Cao, Credit debt default research based on explainable machine learning, (in Chinese), Scientific Decision Making, no. 5, pp. 190–200, 2023.
Google Scholar
[32]

I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, Generative adversarial networks, Commun. ACM, vol. 63, no. 11, pp. 139–144, 2020.
DOI Google Scholar
[33]

K. W. Sun and C. H. Lee, Addressing class-imbalance in multi-label learning via two-stage multi-label hypernetwork, Neurocomputing, vol. 266, no. C, pp. 375–389, 2017.
DOI Google Scholar
[34]

C. Zhang, F. Zhang, N. Chen, and H. Long, Application of artificial intelligence technology in financial data inspection and manufacturing bond default prediction in small and medium-sized enterprises (SMEs), Oper. Manag. Res., vol. 15, no. 3, pp. 941–952, 2022.
DOI Google Scholar
[35]
A. A. M. Al-Saffar, H. Tao, and M. A. Talab, Review of deep convolution neural network in image classification, in Proc. Int. Conf. Radar, Antenna, Microwave, Electronics, and Telecommunications (ICRAMET), Jakarta, Indonesia, 2017, pp. 26–31.
DOI
[36]

J. Zhai, J. Qi, and S. Zhang, Imbalanced data classification based on diverse sample generation and classifier fusion, Int. J. Mach. Learn. Cybern., vol. 13, no. 3, pp. 735–750, 2022.
DOI Google Scholar
[37]

M. Kizildag and O. Ozdemir, Underlying factors of ups and downs in financial leverage overtime, Tour. Econ., vol. 23, no. 6, pp. 1321–1342, 2017.
DOI Google Scholar
[38]

J. Piñeiro-Chousa, M. Á. López-Cabarcos, J. Caby, and A. Šević, The influence of investor sentiment on the green bond market, Technol. Forecast. Soc. Change, vol. 162, p. 120351, 2021.
DOI Google Scholar
Journal of Social Computing
Volume 5 Issue 1,
March 2024
Pages 36-45
DOI: 10.23919/JSC.2024.0005
Cite this article:
Zhang N, Li W, Chen H, et al. Predicting Credit Bond Default with Deep Learning: Evidence from China. Journal of Social Computing, 2024, 5(1): 36-45. https://doi.org/10.23919/JSC.2024.0005

111

Views

16

Downloads

0

Crossref

0

Scopus

Altmetrics
Received: 27 October 2023
Revised: 06 March 2024
Accepted: 08 March 2024
Published: 30 March 2024
© The author(s) 2024.

The articles published in this open access journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).
Return