Comparison of Credit Scoring Models on Probability of Default Estimation for Us Banks

Gurný, Petr; Gurny, Martin

doi:10.18267/j.pep.446

Cited by 29 publications

(21 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The probabilities it outputs allow for great flexibility regarding an acceptable risk threshold (Kostin, 2018), which can be derived empirically and set manually. In several studies, LR outperformed models based on Random Forrest (Abdemoula, 2015;Çığşar & Ünal, 2019), Naïve Bayes (Abdemoula, 2015;Çığşar & Ünal, 2019), K Nearest Neighbor (KNN) (Abdemoula, 2015) and discriminant analysis or probit (Gurný & Gurný, 2013).…”

Section: Logistic Regressionmentioning

confidence: 99%

“…This indicates increased interest for P2P platforms in academia. Based on data from banks, ten studies focused on private individuals (PI) (Akcura & Chhibber, 2018;Bayraci & Susuz, 2019;Çığşar & Ünal, 2019;Gurný & Gurný, 2013;Kaya, Gurgen, & Okay, 2008;Nabende & Senfuma, 2019;Nh & Mai, 2018;Obare & Muraya, 2018;Sariev & Germano, 2020;Turlík, 2018) and seven on companies (Abdemoula, 2015;Härdle, Moro, & Schäfer, 2007;Ploeg, Verschoor, & Menken, 2010;Raei, Kousha, Fallahpour, & Fadacinejad, 2016;Ramakrishnan, Mirzaei, & Bekri, 2015;Sariev & Germano, 2020;Wang, Cao, Lu, & Wang, 2013).…”

Section: Applicationsmentioning

confidence: 99%

“…The models that were found best most often were hybrid ones ( (Kaya, Gurgen, & Okay, 2008;Raei, Kousha, Fallahpour, & Fadacinejad, 2016;Turiel & Aste, 2020;Wang, Cao, Lu, & Wang, 2013), followed by LR (Abdemoula, 2015;Çığşar & Ünal, 2019;Gurný & Gurný, 2013) and SVM (Akcura & Chhibber, 2018;Härdle, Moro, & Schäfer, 2007;Obare & Muraya, 2018). Hybrid models proved most performant for companies taking loans from banks, while for bank PI customers, most often the most accurate models were LR, NN and SVM.…”

Section: Applicationsmentioning

confidence: 99%

“…Firstly, best models in traditional banking tend to form clusters regarding the population they are used on. For example, logistic regression (Çığşar & Ünal, 2019;Gurný & Gurný, 2013), neural networks (Bayraci & Susuz, 2019;Sariev & Germano, 2020) and SVM (Akcura & Chhibber, 2018;Obare & Muraya, 2018)models seem to perform best for private individual credit scoring models, while the SME/Corporate sectors benefit more from hybrid models (Raei, Kousha, Fallahpour, & Fadacinejad, 2016;Wang, Cao, Lu, & Wang, 2013). Admittedly, most of these models presented in the literature are not necessarily implemented in the national context they were developed for.…”

Section: Limitations and Future Researchmentioning

confidence: 99%

See 3 more Smart Citations

Will they repay their debt? Identification of borrowers likely to be charged off

Căplescu

Panaite

Pele

et al. 2020

Management &Amp; Marketing. Challenges for the Knowledge Society

View full text Add to dashboard Cite

Recent increase in peer-to-peer lending prompted for development of models to separate good and bad clients to mitigate risks both for lenders and for the platforms. The rapidly increasing body of literature provides several comparisons between various models. Among the most frequently employed ones are logistic regression, Support Vector Machines, neural networks and decision tree-based models. Among them, logistic regression has proved to be a strong candidate both because its good performance and due to its high explainability. The present paper aims to compare four pairs of models (for imbalanced and under-sampled data) meant to predict charged off clients by optimizing F1 score. We found that, if the data is balanced, Logistic Regression, both simple and with Stochastic Gradient Descent, outperforms LightGBM and K-Nearest Neighbors in optimizing F1 score. We chose this metric as it provides balance between the interests of the lenders and those of the platform. Loan term, debt-to-income ratio and number of accounts were found to be important positively related predictors of risk of charge off. At the other end of the spectrum, by far the strongest impact on charge off probability is that of the FICO score. The final number of features retained by the two models differs very much, because, although both models use Lasso for feature selection, Stochastic Gradient Descent Logistic Regression uses a stronger regularization. The analysis was performed using Python (numpy, pandas, sklearn and imblearn).

show abstract

Section: Logistic Regressionmentioning

confidence: 99%

Section: Applicationsmentioning

confidence: 99%

Section: Applicationsmentioning

confidence: 99%

Section: Limitations and Future Researchmentioning

confidence: 99%

See 2 more Smart Citations

Will they repay their debt? Identification of borrowers likely to be charged off

Căplescu

Panaite

Pele

et al. 2020

Management &Amp; Marketing. Challenges for the Knowledge Society

View full text Add to dashboard Cite

show abstract

“…The empirical results obtained show significant links between credit risk and variables such as ROE and ROA, inefficiency (IE), loan-to-deposit ratio (LDR), credit growth (CG) and deposit rate (DR); at the same time, variables of solvency (SR) and credit rate (CR) are not statistically significant in terms of credit risk. Gurný and Gurný (2013) assessed the default probability of American banks through statistical models of credit scoring. Given the considered models with the use of linear discriminant analysis and regression models and taking into account the statistical significance of the estimated parameters, the authors analyzed the sample of three hundred US commercial banks, which are divided into two groups (with and without default) based on historical information.…”

Section: Literature Reviewmentioning

confidence: 99%

Defining the probability of bank debtors’ default using financial solvency assessment models

Kuznichenko¹,

Dykha²,

Павлова³

et al. 2018

Banks and Bank Systems

View full text Add to dashboard Cite

Due implementation of debtors’ financial solvency assessment models by Ukrainian banks with the aim of calculating the probability of their default (PD) is the next step towards the integration of Ukrainian banking system into global banking community, convergence of methodical approaches to assessing the credit risk with standards of international practice, possibility of using IRB-approach (an approach based on internal ratings) for calculating the regulatory requirements to capital adequacy.The analysis of approaches to bank credit portfolio segmentation according to types of debtors and debtors’ financial solvency assessment models, depending on the performed segmentation and accumulated bank statistical data, from the point of view of its suitability for Ukrainian banks, will enable the banks to choose the most suitable ones for implementation taking into account nature and complexity of operations performed.Such approaches will be more adapted to minimum capital requirements, simultaneously agreeing with national supervisory priorities.

show abstract

The weakest links in the crisis of the Spanish Savings Banks

2022

View full text Add to dashboard Cite

The bail-out of the Spanish financial system was the largest in the Eurozone. The Spanish savings banks were the main direct recipients of public support between 2010 and 2012 and most of the bail-out money was transferred to the savings banks after a series of mergers and acquisitions were completed among these entities. This article analyses the evolution of the savings banks in the years of economic growth preceding the crisis and their performance in the early aftermath of the 2007 crash using previously unexplored data from the annual financial statements of the 45 savings banks. We conclude it is possible to observe distinct behaviours among the savings banks during the boom years and we estimate the largest part of the bail-out money corresponds to losses registered by a small number of savings banks that were already severely impaired before the merger process got under way.

show abstract

Comparison of Credit Scoring Models on Probability of Default Estimation for Us Banks

Cited by 29 publications

References 10 publications

Will they repay their debt? Identification of borrowers likely to be charged off

Will they repay their debt? Identification of borrowers likely to be charged off

Defining the probability of bank debtors’ default using financial solvency assessment models

The weakest links in the crisis of the Spanish Savings Banks

Contact Info

Product

Resources

About