Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box Approach
Predicting bank failures is a critical task requiring balancing the need for model explainability with the necessity of preserving data privacy. Traditional machine learning models often lack transparency, which poses challenges for stakeholders who need to understand the factors leading to predicti...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10818483/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825207008295387136 |
|---|---|
| author | Junyoung Byun Jaewook Lee Hyeongyeong Lee Bumho Son |
| author_facet | Junyoung Byun Jaewook Lee Hyeongyeong Lee Bumho Son |
| author_sort | Junyoung Byun |
| collection | DOAJ |
| description | Predicting bank failures is a critical task requiring balancing the need for model explainability with the necessity of preserving data privacy. Traditional machine learning models often lack transparency, which poses challenges for stakeholders who need to understand the factors leading to predictions. In this study, we employ differentially private glass-box models, namely Explainable Boosting Machine (EBM) and Neural Additive Models (NAM), to address these issues. We analyzed data from 21,243 American banks spanning from 1969 to 2021, focusing on key financial ratios. By applying Differential Privacy (DP) to these models, we aimed to protect sensitive financial data while evaluating the trade-offs between privacy, accuracy, and explainability. Our main findings are as follows: 1) In the absence of privacy constraints, the models consistently identified Asset Turnover, Total Debt / Invested Capital, and ROE ratios as the most influential factors in predicting bank failure, in that order; 2) When the privacy budget <inline-formula> <tex-math notation="LaTeX">$\epsilon \leq 1$ </tex-math></inline-formula>, only EBM maintained significant performance; 3) The reduction in explainability due to privacy protection was more pronounced for variables with initially lower explanatory power, while Asset Turnover retained its explanatory power even at <inline-formula> <tex-math notation="LaTeX">$\epsilon = 0.01$ </tex-math></inline-formula>. These findings provide valuable insights for banks, policymakers, and investors, suggesting that glass-box models can offer a promising solution for reliable and explainable bank failure prediction under privacy constraints. |
| format | Article |
| id | doaj-art-3105c36b10ce4536a2162ee97d5ccf7c |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-3105c36b10ce4536a2162ee97d5ccf7c2025-02-07T00:01:54ZengIEEEIEEE Access2169-35362025-01-01131546156510.1109/ACCESS.2024.352396710818483Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box ApproachJunyoung Byun0Jaewook Lee1https://orcid.org/0000-0001-5720-8337Hyeongyeong Lee2Bumho Son3https://orcid.org/0000-0003-0856-7039Department of Applied Statistics, Chung-Ang University, Seoul, Republic of KoreaDepartment of Industrial Engineering, Seoul National University, Seoul, Republic of KoreaDepartment of Applied Statistics, Chung-Ang University, Seoul, Republic of KoreaDepartment of Business Administration, Chung-Ang University, Seoul, Republic of KoreaPredicting bank failures is a critical task requiring balancing the need for model explainability with the necessity of preserving data privacy. Traditional machine learning models often lack transparency, which poses challenges for stakeholders who need to understand the factors leading to predictions. In this study, we employ differentially private glass-box models, namely Explainable Boosting Machine (EBM) and Neural Additive Models (NAM), to address these issues. We analyzed data from 21,243 American banks spanning from 1969 to 2021, focusing on key financial ratios. By applying Differential Privacy (DP) to these models, we aimed to protect sensitive financial data while evaluating the trade-offs between privacy, accuracy, and explainability. Our main findings are as follows: 1) In the absence of privacy constraints, the models consistently identified Asset Turnover, Total Debt / Invested Capital, and ROE ratios as the most influential factors in predicting bank failure, in that order; 2) When the privacy budget <inline-formula> <tex-math notation="LaTeX">$\epsilon \leq 1$ </tex-math></inline-formula>, only EBM maintained significant performance; 3) The reduction in explainability due to privacy protection was more pronounced for variables with initially lower explanatory power, while Asset Turnover retained its explanatory power even at <inline-formula> <tex-math notation="LaTeX">$\epsilon = 0.01$ </tex-math></inline-formula>. These findings provide valuable insights for banks, policymakers, and investors, suggesting that glass-box models can offer a promising solution for reliable and explainable bank failure prediction under privacy constraints.https://ieeexplore.ieee.org/document/10818483/Bank failure predictiondifferential privacyexplainable artificial intelligenceexplainable boosting machineneural additive models |
| spellingShingle | Junyoung Byun Jaewook Lee Hyeongyeong Lee Bumho Son Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box Approach IEEE Access Bank failure prediction differential privacy explainable artificial intelligence explainable boosting machine neural additive models |
| title | Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box Approach |
| title_full | Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box Approach |
| title_fullStr | Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box Approach |
| title_full_unstemmed | Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box Approach |
| title_short | Balancing Explainability and Privacy in Bank Failure Prediction: A Differentially Private Glass-Box Approach |
| title_sort | balancing explainability and privacy in bank failure prediction a differentially private glass box approach |
| topic | Bank failure prediction differential privacy explainable artificial intelligence explainable boosting machine neural additive models |
| url | https://ieeexplore.ieee.org/document/10818483/ |
| work_keys_str_mv | AT junyoungbyun balancingexplainabilityandprivacyinbankfailurepredictionadifferentiallyprivateglassboxapproach AT jaewooklee balancingexplainabilityandprivacyinbankfailurepredictionadifferentiallyprivateglassboxapproach AT hyeongyeonglee balancingexplainabilityandprivacyinbankfailurepredictionadifferentiallyprivateglassboxapproach AT bumhoson balancingexplainabilityandprivacyinbankfailurepredictionadifferentiallyprivateglassboxapproach |