SHAP Stability in Credit Risk Management: A Case Study in Credit Card Default Model

Paper: arXiv 2508.01851

Authors: Luyun Lin, Yiqing Wang

Abstract

The increasing development in the consumer credit card market brings substantial regulatory and risk management challenges. The advanced machine learning models applications bring concerns about model transparency and fairness for both financial institutions and regulatory departments. In this study, we evaluate the consistency of one commonly used Explainable AI (XAI) technology, SHAP, for variable explanation in credit card probability of default models via a case study about credit card default prediction. The study shows the consistency is related to the variable importance level and hence provides practical recommendation for credit risk management

Complexity vs Empirical Score

• Math Complexity: 2.5/10
• Empirical Rigor: 6.5/10
• Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper presents a practical empirical study evaluating the stability of SHAP values across multiple XGBoost model runs on a standard benchmark credit dataset. While the mathematical rigor is low and relies primarily on standard introductory concepts (KS statistic, confusion matrix), the empirical setup is well-designed to address a critical regulatory compliance question in credit risk management.

Research Flowchart

  flowchart TD
    A[Research Goal: Evaluate SHAP Consistency in Credit Default Models] --> B[Data Input: Credit Card Default Dataset]
    B --> C[Train Machine Learning Models for Probability of Default]
    C --> D[Apply SHAP to Calculate Feature Importance & Explanations]
    D --> E[Assess SHAP Explanation Stability Across Variables]
    E --> F[Findings: SHAP Consistency Correlates with Variable Importance Level]
    F --> G[Practical Recommendations for Credit Risk Management & XAI]