A survey of machine learning in credit risk

Breeden, Joseph L.

doi:10.21314/jcr.2021.008

Cited by 17 publications

(14 citation statements)

References 134 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning models require more data for training to avoid overfitting 39 which is assured by introducing the Kaggle and Microcredit datasets. Fourth, mostly applied credit scoring data sets are not entirely appropriate for deep learning state‐of‐the‐art algorithms, which reach their full potential only by using larger data sets 9 …”

Section: Datasets and Methodologymentioning

confidence: 99%

“…Therefore, an important research problem is to further improve the performance of credit scoring based on deep learning technology. 5,13 One of the newer surveys on deep learning applications 9 does not mention any of the concrete state-of-the-art deep learning algorithms applied to credit scoring datasets. According to the review, 6 A recent study 5 has applied an ensemble model with LSTM as a base technique on classical credit scoring datasets, that is, on German and Taiwan datasets.…”

Section: Lstm Techniques In Credit Scoringmentioning

confidence: 99%

“…Therefore, an important research problem is to further improve the performance of credit scoring based on deep learning technology 5,13 . One of the newer surveys on deep learning applications 9 does not mention any of the concrete state‐of‐the‐art deep learning algorithms applied to credit scoring datasets. According to the review, 6 none of the studies have covered the LSTM deep learning technique on credit scoring models, but it is evident that a considerable number of papers that cover LSTM other financial fields such as algo‐trading, fraud detection, financial distress, bankruptcy, bank risk, mortgage risk, crisis forecasting studies, portfolio management, asset pricing, and derivatives market studies, crypto currency, and block chain studies, and financial sentiment studies coupled with text mining for forecasting.…”

Section: Literature Reviewmentioning

confidence: 99%

“…In this article datasets of different sizes and origins, with different numbers and types of attributes are used: German, Kaggle, and Microcredit datasets. The absurdity was that most of the published works used big data machine learning methods on extremely small datasets 9 . It is not good practice to conclude that one method is better than another based only on one dataset over one period 9 …”

Section: Introductionmentioning

confidence: 99%

“…The absurdity was that most of the published works used big data machine learning methods on extremely small datasets 9 . It is not good practice to conclude that one method is better than another based only on one dataset over one period 9 Another contribution is to overcome the limitations of public datasets that the studies rely on.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Intelligent credit scoring using deep learning methods

Gicić

Ðonko

Subaşı

2023

Concurrency and Computation

View full text Add to dashboard Cite

Summary Credit scoring is one the most important parts of credit risk management in reducing the risk of client defaults and bankruptcies. Deep learning has received much attention in recent years, but it has not been implemented so intensively in credit scoring compared to other financial domains. In this article, stacked unidirectional and bidirectional LSTM (long short‐term memory) networks as a complex area of deep learning are applied in solving credit scoring problems for the first time. The proposed robust model exploits the full potential of the three‐layer stacked LSTM and BDLSTM (bidirectional LSTM) architecture with the treatment and modeling of public datasets in a novel way since credit scoring is not a time sequence problem. Attributes of each loan instance were transformed into a sequence of the matrix with a fixed sliding window approach with a one‐time step. Our proposed models outperform existing and much more complex deep learning solutions thus we succeeded in preserving simplicity. In this article, measures of different types are employed to carry out consistent conclusions. The results by applying three hidden layers on the German Credit dataset showed an accuracy of 87.19%, for Kaggle dataset accuracy reached 93.69%, and for Microcredit dataset accuracy of 97.80%.

show abstract

Section: Datasets and Methodologymentioning

confidence: 99%

Section: Lstm Techniques In Credit Scoringmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Intelligent credit scoring using deep learning methods

Gicić

Ðonko

Subaşı

2023

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

Lost in a black‐box? Interpretable machine learning for assessing Italian SMEs default

Crosato,

Liberati,

Repetto

2023

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

Academic research and the financial industry have recently shown great interest in Machine Learning algorithms capable of solving complex learning tasks, although in the field of firms' default prediction the lack of interpretability has prevented an extensive adoption of the black‐box type of models. In order to overcome this drawback and maintain the high performances of black‐boxes, this paper has chosen a model‐agnostic approach. Accumulated Local Effects and Shapley values are used to shape the predictors' impact on the likelihood of default and rank them according to their contribution to the model outcome. Prediction is achieved by two Machine Learning algorithms (eXtreme Gradient Boosting and FeedForward Neural Networks) compared with three standard discriminant models. Results show that our analysis of the Italian Small and Medium Enterprises manufacturing industry benefits from the overall highest classification power by the eXtreme Gradient Boosting algorithm still maintaining a rich interpretation framework to support decisions.

show abstract

Can we trust machine learning to predict the credit risk of small businesses?

Bitetto,

Cerchiello,

Filomeni

et al. 2024

Rev Quant Finan Acc

View full text Add to dashboard Cite

With the emergence of Fintech lending, small firms can benefit from new channels of financing. In this setting, the creditworthiness and the decision to extend credit are often based on standardized and advanced machine-learning techniques that employ limited information. This paper investigates the ability of machine learning to correctly predict credit risk ratings for small firms. By employing a unique proprietary dataset on invoice lending activities, this paper shows that machine learning techniques overperform traditional techniques, such as probit, when the set of information available to lenders is limited. This paper contributes to the understanding of the reliability of advanced credit scoring techniques in the lending process to small businesses, making it a special interesting case for the Fintech environment.

show abstract

A survey of machine learning in credit risk

Cited by 17 publications

References 134 publications

Intelligent credit scoring using deep learning methods

Intelligent credit scoring using deep learning methods

Lost in a black‐box? Interpretable machine learning for assessing Italian SMEs default

Can we trust machine learning to predict the credit risk of small businesses?

Contact Info

Product

Resources

About