Loan lending plays an important role in our everyday life and powerfully promotes the growth of consumption and the economy. Loan default has been unavoidable, which carries a great risk and may even end up in a finan...Loan lending plays an important role in our everyday life and powerfully promotes the growth of consumption and the economy. Loan default has been unavoidable, which carries a great risk and may even end up in a financial crisis. Therefore, it is particularly important to identify whether a candidate is eligible for receiving a loan. In this paper, we apply Random Forest and XGBoost algorithms to train the prediction model and compare their performance in prediction accuracy. In the feature engineering part, we use the variance threshold method and Variance Inflation Factor method to filter out unimportant features, and then we input those selected features into Random Forest and XGBoost models. It turns out that Random Forest and XGBoost show little difference in the accuracy of their predictions since both get high accuracy of around 0.9 in the loan default cases.展开更多
Owing to the convenience of online loans,an increasing number of people are borrowing money on online platforms.With the emergence of machine learning technology,predicting loan defaults has become a popular topic.How...Owing to the convenience of online loans,an increasing number of people are borrowing money on online platforms.With the emergence of machine learning technology,predicting loan defaults has become a popular topic.However,machine learning models have a black-box problem that cannot be disregarded.To make the prediction model rules more understandable and thereby increase the user’s faith in the model,an explanatory model must be used.Logistic regression,decision tree,XGBoost,and LightGBM models are employed to predict a loan default.The prediction results show that LightGBM and XGBoost outperform logistic regression and decision tree models in terms of the predictive ability.The area under curve for LightGBM is 0.7213.The accuracies of LightGBM and XGBoost exceed 0.8.The precisions of LightGBM and XGBoost exceed 0.55.Simultaneously,we employed the local interpretable model-agnostic explanations approach to undertake an explainable analysis of the prediction findings.The results show that factors such as the loan term,loan grade,credit rating,and loan amount affect the predicted outcomes.展开更多
Between 2005 and 2007,the China Development Bank offered 1.66 billion yuan($237 million) worth of loans to 243,000 stu- dents from poor families in central China’s Henan Province.
In this study,we use bank loan information to construct proxies for corporate transparency and examine whether these measures reflect information asymmetry in the stock market.Our analysis is based on a novel dataset ...In this study,we use bank loan information to construct proxies for corporate transparency and examine whether these measures reflect information asymmetry in the stock market.Our analysis is based on a novel dataset of stock transactions and bank loans of all publicly listed firms on the Shenzhen Stock Exchange,covering January 2008 to June 2013.We find that firms with outstanding loans have a lower level of information asymmetry in the stock market,whereas firms with defaulted loans have a higher level of asymmetry.Further evidence demonstrates that the effect of loan default on information asymmetry in the stock market is more pronounced when these loans are borrowed from joint-equity commercial banks or multiple banks and when the default occurs under inactive market conditions.Our results remain robust to a series of endogeneity and sensitivity tests and provide suggestive evidence of a close connection between the credit loan and stock markets.展开更多
The purpose of this paper is to come up with factors in loan loss provisioning practices on commercial banks that reflect on collectability of defaulted loans. The need for this research is due to failures in the loan...The purpose of this paper is to come up with factors in loan loss provisioning practices on commercial banks that reflect on collectability of defaulted loans. The need for this research is due to failures in the loan loss provisioning practices which resulted in loan loss provisions （LLP） not reflecting on collectability of the defaulted loans. As a consequence, the banks do not capture their loss expectations and do not continuously reassess their loss expectations as the conditions affecting their borrowers may change. Henceforth, in their financial reporting, the banks do not represent relevantly and faithfully their true underlying credit risks conditions. When the banks do not represent relevantly and faithfully their true underlying risk conditions, they contradict the objectives of useful financial reporting. The results showed that among explanatory variables, bad debt recoveries as a factor in loan loss provisioning practices that reflect on collectability of defaulted loans was rejected. Bad debt recoveries was a biased variable and inconsistent estimator. In context of perceived credit risks as the basis to make credit judgments, an estimate of bad debt recoveries had not fulfilled the criteria. On the other hand, non-performing loans （NPL） as a factor in loan loss provisioning practices was not rejected.展开更多
文摘Loan lending plays an important role in our everyday life and powerfully promotes the growth of consumption and the economy. Loan default has been unavoidable, which carries a great risk and may even end up in a financial crisis. Therefore, it is particularly important to identify whether a candidate is eligible for receiving a loan. In this paper, we apply Random Forest and XGBoost algorithms to train the prediction model and compare their performance in prediction accuracy. In the feature engineering part, we use the variance threshold method and Variance Inflation Factor method to filter out unimportant features, and then we input those selected features into Random Forest and XGBoost models. It turns out that Random Forest and XGBoost show little difference in the accuracy of their predictions since both get high accuracy of around 0.9 in the loan default cases.
基金supported by Fundamental Research Funds for the Central Universities(WUT:2022IVA067).
文摘Owing to the convenience of online loans,an increasing number of people are borrowing money on online platforms.With the emergence of machine learning technology,predicting loan defaults has become a popular topic.However,machine learning models have a black-box problem that cannot be disregarded.To make the prediction model rules more understandable and thereby increase the user’s faith in the model,an explanatory model must be used.Logistic regression,decision tree,XGBoost,and LightGBM models are employed to predict a loan default.The prediction results show that LightGBM and XGBoost outperform logistic regression and decision tree models in terms of the predictive ability.The area under curve for LightGBM is 0.7213.The accuracies of LightGBM and XGBoost exceed 0.8.The precisions of LightGBM and XGBoost exceed 0.55.Simultaneously,we employed the local interpretable model-agnostic explanations approach to undertake an explainable analysis of the prediction findings.The results show that factors such as the loan term,loan grade,credit rating,and loan amount affect the predicted outcomes.
文摘Between 2005 and 2007,the China Development Bank offered 1.66 billion yuan($237 million) worth of loans to 243,000 stu- dents from poor families in central China’s Henan Province.
基金supported by grants from the National Natural Science Foundation of China(72103017,72192800)Fundamental Research Funds for the Central Universities(ZY2130)+1 种基金Funds for First-class Discipline Construction(XK1802-5)“the Fundamental Research Funds for the Central Universities”in UIBE(17DQ08).
文摘In this study,we use bank loan information to construct proxies for corporate transparency and examine whether these measures reflect information asymmetry in the stock market.Our analysis is based on a novel dataset of stock transactions and bank loans of all publicly listed firms on the Shenzhen Stock Exchange,covering January 2008 to June 2013.We find that firms with outstanding loans have a lower level of information asymmetry in the stock market,whereas firms with defaulted loans have a higher level of asymmetry.Further evidence demonstrates that the effect of loan default on information asymmetry in the stock market is more pronounced when these loans are borrowed from joint-equity commercial banks or multiple banks and when the default occurs under inactive market conditions.Our results remain robust to a series of endogeneity and sensitivity tests and provide suggestive evidence of a close connection between the credit loan and stock markets.
文摘The purpose of this paper is to come up with factors in loan loss provisioning practices on commercial banks that reflect on collectability of defaulted loans. The need for this research is due to failures in the loan loss provisioning practices which resulted in loan loss provisions （LLP） not reflecting on collectability of the defaulted loans. As a consequence, the banks do not capture their loss expectations and do not continuously reassess their loss expectations as the conditions affecting their borrowers may change. Henceforth, in their financial reporting, the banks do not represent relevantly and faithfully their true underlying credit risks conditions. When the banks do not represent relevantly and faithfully their true underlying risk conditions, they contradict the objectives of useful financial reporting. The results showed that among explanatory variables, bad debt recoveries as a factor in loan loss provisioning practices that reflect on collectability of defaulted loans was rejected. Bad debt recoveries was a biased variable and inconsistent estimator. In context of perceived credit risks as the basis to make credit judgments, an estimate of bad debt recoveries had not fulfilled the criteria. On the other hand, non-performing loans （NPL） as a factor in loan loss provisioning practices was not rejected.
