Paper: arXiv 2401.09778

Authors: O. Didkovskyi, N. Jean, G. Le Pera, C. Nordio

Abstract

This paper introduces a credit risk rating model for credit risk assessment in quantitative finance, aiming to categorize borrowers based on their behavioral data. The model is trained on data from Experian, a widely recognized credit bureau, to effectively identify instances of loan defaults among bank customers. Employing state-of-the-art statistical and machine learning techniques ensures the model’s predictive accuracy. Furthermore, we assess the model’s transferability by testing it on behavioral data from the Bank of Italy, demonstrating its potential applicability across diverse datasets during prediction. This study highlights the benefits of incorporating external behavioral data to improve credit risk assessment in financial institutions.

Complexity vs Empirical Score

  • Math Complexity: 4.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper employs standard statistical and machine learning methods (Logistic Regression, AutoML, Genetic Algorithms, SHAP) without heavy theoretical derivations, keeping math complexity moderate. It demonstrates high empirical rigor through rigorous data filtering, feature engineering, out-of-sample/out-of-time validation, probability calibration, and backtesting on an external central credit register dataset.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Develop a credit risk model<br>for behavioral data"] --> B["Data Source: Experian<br>(Private Data)"]
  A --> C["Data Source: Bank of Italy<br>(Central Data)"]
  B --> D["Model Training & Validation<br>Using ML Techniques"]
  C --> E["Transferability Test<br>Unseen Data Evaluation"]
  D --> F["Key Finding: High Predictive<br>Accuracy on Experian"]
  E --> G["Key Finding: Successful<br>Transferability Across Datasets"]
  F --> H["Outcome:<br>Enhanced Credit Risk<br>Assessment Framework"]
  G --> H

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a credit risk rating model trained on Experian data and tests its transferability to behavioral data from the Bank of Italy.

Source sentence
This paper introduces a credit risk rating model for credit risk assessment in quantitative finance, aiming to categorize borrowers based on their behavioral data.

Assumes 5

  • Experian data are a subset of the Central Credit Register (CR) data. data
    Source sentence
    Each adjusted p-value is less than 0.0001, hence all null hypotheses are rejected. This outcome affirms that Experian data are a subset of CR.
  • The mapping between CR and Experian features is correct and can be applied at prediction time. model
    Source sentence
    To ensure the correctness of the mapping we set up a rigorous dual-strategy approach.
  • Active companies without any “insolvency” status represent a stable dataset for modeling. data
    Source sentence
    Our analysis specifically focused on active companies that did not have any records of “insolvency” as their worst special status, ensuring a more stable dataset.
  • Features with more than 20% missing values are removed because they are likely to be missing at prediction. model
    Source sentence
    The initial and most straightforward data cleaning step involves eliminating features that contain more than 20% missing values (NaNs).
Show 1 more
  • Out‑of‑sample (80/20) and out‑of‑time (2018‑2021 vs 2022) splits provide valid validation of model performance. statistical
    Source sentence
    Upon data filtering, we implemented out-of-sample and out-of-time validation, splitting the dataset twice: 80%-20% for (stratified) out-of-sample, and datasets from 2018-2021 versus 2022 for out-of-time.

Data

  • Experian dataset · Experian alternative commercial
    2018, 2019, 2020, 2021, 2022 · Corporate clients (active companies)
    Source sentence
    An extensive dataset was created by collecting data from five different time periods: March 2018, 2019, 2020, 2021, and 2022.
  • Central Credit Register (CR) · Bank of Italy alternative proprietary
    National banking system corporate borrowers
    Source sentence
    We introduce a potential feature mapping technique to enable the utilization of the calibrated model with publicly accessible data from the Central Credit Register.

Costs

Not applicable The paper reports no trading or execution result.

Finds 2

  • The Light‑GBM classifier achieves an AUC of approximately 0.92‑0.93 in both out‑of‑sample and out‑of‑time validation. predictive-accuracy
    Source sentence
    Specifically, it achieves an AUC of approximately 0.92 in both out-of-sample and out-of-time validation, as depicted in Figure 3.
  • Baseline Experian DG3r score yields AUC values around 0.86 across the five datasets. predictive-accuracy
    Source sentence
    For instance, DG3r’s AUC values for each of the five datasets are approximately 0.86.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 18 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (54)

  1. Isolation forest (2008)
  2. Credit Research Database
  3. Global Economic Databank
  4. Non-performing loans (NPLs) in Italy’s banking system
  5. Applied regression analysis (1998)
  6. Hands-on machine learning with Scikit-Learn and TensorFlow: concepts, tools, and techniques to build intelligent systems (2017)
  7. Quantitative risk management: Concepts, techniques and tools (2005)
  8. Entity embeddings of categorical variables (2016) — link
and 46 more
  1. An empirical analysis of feature engineering for predictive modeling (2016)
  2. Feature engineering for machine learning: principles and techniques for data scientists (2018)
  3. A unified approach to interpreting model predictions (2017)
  4. " Why should i trust you?" Explaining the predictions of any classifier (2016)
  5. Reconsidering ratings (2005)
  6. A traffic lights approach to PD validation (2003)
  7. Dynamic programming: Princeton univ. press (1957)
  8. Neural network methods for natural language processing (2017)
  9. A neural probabilistic language model (2003)
  10. Practical lessons from predicting clicks on ads at facebook (2014)
  11. Stochastic gradient boosting (2002)
  12. Differential Evolution - A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces (1997) — link
  13. Matching industry classifications. A method for converting Nace Rev. 2 to Nace Rev. 1 (2015)
  14. Focal loss for dense object detection (2017)
  15. Forecasting US bond default ratings allowing for previous and initial state dependence in an ordered probit model (2012)
  16. Comparison of credit scoring models on probability of default estimation for us banks (2013)
  17. Investigation and improvement of multi-layer perceptron neural networks for credit scoring (2015)
  18. Credit risk analysis using machine and deep learning models (2018)
  19. A robust machine learning approach for credit risk analysis of large loan level datasets using deep learning and extreme gradient boosting (2019)
  20. Autogluon-tabular: Robust and accurate automl for structured data (2020) — link
  21. Machine learning approach for credit scoring (2020) — link
  22. An Artificial Intelligence approach to Shadow Rating (2019) — link
  23. Predicting consumer default: A deep learning approach (2019)
  24. Consumer credit-risk models via machine-learning algorithms (2010)
  25. Deep Learning for Mortgage Risk (2018)
  26. Risk and risk management in the credit card industry (2016)
  27. Forecasting bankruptcy for SMEs using hazard function: To what extent does size matter? (2015)
  28. A quantitative theory of unsecured consumer credit with risk of default (2007)
  29. Feature contribution alignment with expert knowledge for artificial intelligence credit scoring (2023)
  30. Explainable machine learning in credit risk management (2021)
  31. Lightgbm: A highly efficient gradient boosting decision tree (2017)
  32. The elements of statistical learning: data mining, inference, and prediction (2009)
  33. Optuna: A next-generation hyperparameter optimization framework (2019)
  34. A note on the validity of cross-validation for evaluating autoregressive time series prediction (2018)
  35. Obtaining Calibrated Probabilities from Boosting (2012) — link
  36. Strictly Proper Scoring Rules, Prediction, and Estimation (2007) — link
  37. Beta calibration: a well-founded and easily implemented improvement on logistic calibration for binary classifiers (2017)
  38. Weighted kappa: nominal scale agreement provision for scaled disagreement or partial credit. (1968)
  39. The Kendall rank correlation coefficient (2007)
  40. A similarity measure for indefinite rankings (2010)
  41. Feature selection with the Boruta package (2010)
  42. Circular no. 139. Central Credit Register (1991) — link
  43. Controlling the false discovery rate: a practical and powerful approach to multiple testing (1995)
  44. The control of the false discovery rate in multiple testing under dependency (2001)
  45. Practical nonparametric statistics (1999)
  46. Note on the sampling error of the difference between correlated proportions or percentages (1947)

0 of 54 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.