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 mark...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.展开更多
A credit risk prediction model named KM-ADASYN-TL-FLLightGBM(KADT-FLightGBM)is proposed in this study.Firstly,to overcome the limitation of traditional sampling methods in dealing with imbalanced datasets,an improved ...A credit risk prediction model named KM-ADASYN-TL-FLLightGBM(KADT-FLightGBM)is proposed in this study.Firstly,to overcome the limitation of traditional sampling methods in dealing with imbalanced datasets,an improved ADASYN sampling with K-means clustering algorithm is constructed.Moreover,the Tomek Links method is used to filter the generated samples.Secondly,an utilized an optimized LightGBM algorithm with the Focal Loss is employed to training the model using the datasets obtained by the improved ADASYN sampling.Finally,the comparative analysis between the ensemble model and other different sampling methodologies is conducted on the Lending Club dataset.The results demonstrate that the proposed model effectively minimizes the misclassification of minority classes in credit risk prediction and can be used as a reference for similar studies.展开更多
Credit risk prediction models seek to predict quality factors such as whether an individual will default （bad applicant） on a loan or not （good applicant）. This can be treated as a kind of machine learning （ML） ...Credit risk prediction models seek to predict quality factors such as whether an individual will default （bad applicant） on a loan or not （good applicant）. This can be treated as a kind of machine learning （ML） problem. Recently, the use of ML algorithms has proven to be of great practical value in solving a variety of risk problems including credit risk prediction. One of the most active areas of recent research in ML has been the use of ensemble （combining） classifiers. Research indicates that ensemble individual classifiers lead to a significant improvement in classification performance by having them vote for the most popular class. This paper explores the predicted behaviour of five classifiers for different types of noise in terms of credit risk prediction accuracy, and how could such accuracy be improved by using pairs of classifier ensembles. Benchmarking results on five credit datasets and comparison with the performance of each individual classifier on predictive accuracy at various attribute noise levels are presented. The experimental evaluation shows that the ensemble of classifiers technique has the potential to improve prediction accuracy.展开更多
基金supported in part by the Emerging Interdisciplinary Project of Central University of Finance and Economics,Beijing,China.
文摘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.
基金supported by the National Natural Science Foundation of China(Nos.71503108 and 62077029)CCF-Huawei Innovation Research Program Grant(No.CCF-HuaweiFM202209)Research and Practice Innovation Project of Jiangsu Normal University(No.2022XKT1540).
文摘A credit risk prediction model named KM-ADASYN-TL-FLLightGBM(KADT-FLightGBM)is proposed in this study.Firstly,to overcome the limitation of traditional sampling methods in dealing with imbalanced datasets,an improved ADASYN sampling with K-means clustering algorithm is constructed.Moreover,the Tomek Links method is used to filter the generated samples.Secondly,an utilized an optimized LightGBM algorithm with the Focal Loss is employed to training the model using the datasets obtained by the improved ADASYN sampling.Finally,the comparative analysis between the ensemble model and other different sampling methodologies is conducted on the Lending Club dataset.The results demonstrate that the proposed model effectively minimizes the misclassification of minority classes in credit risk prediction and can be used as a reference for similar studies.
文摘Credit risk prediction models seek to predict quality factors such as whether an individual will default （bad applicant） on a loan or not （good applicant）. This can be treated as a kind of machine learning （ML） problem. Recently, the use of ML algorithms has proven to be of great practical value in solving a variety of risk problems including credit risk prediction. One of the most active areas of recent research in ML has been the use of ensemble （combining） classifiers. Research indicates that ensemble individual classifiers lead to a significant improvement in classification performance by having them vote for the most popular class. This paper explores the predicted behaviour of five classifiers for different types of noise in terms of credit risk prediction accuracy, and how could such accuracy be improved by using pairs of classifier ensembles. Benchmarking results on five credit datasets and comparison with the performance of each individual classifier on predictive accuracy at various attribute noise levels are presented. The experimental evaluation shows that the ensemble of classifiers technique has the potential to improve prediction accuracy.
