UPI behavioral proxies for thin-file credit risk prediction

Beyond Aggregate Calibration: Decomposing Income-Conditional Recall Disparities in Automated Credit Default Prediction

Empirical findings show that simply blinding an algorithm to sensitive attributes fails to ensure fairness when institutional pricing decisions and behavioral proxy variables collectively reconstruct the omitted signals, and outline the practical implications for auditing data-centric AI workflows within regulated financial institutions.

Interpretable hybrid credit scoring for thin-file and underbanked populations

We extend a residual-learning hybrid credit scoring framework (logistic regression scorecard plus a gradient-boosting correction on its residuals, decomposed at each prediction into an interpretability ratio $\rho(x)$ that measures the share attributable to the linear branch) along three axes: an East African empirical instantiation on the Zindi Financial Inclusion in Africa data (Kenya, Rwanda, Tanzania, Uganda); a fairness audit at the granularity of the framework's three interpretability regions; and a thin-file segmentation analysis. On the Taiwan Credit Default benchmark retained for continuity, the calibrated hybrid attains AUC $= 0.776$ ($\Delta\mathrm{AUC} = +0.057$ vs.\ standalone logistic regression, $+0.001$ vs.\ standalone XGBoost), reduces Brier Score by 23\%, and concentrates the highest-default-rate borrowers (69.5\%) in the fully interpretable region. On Zindi, the calibrated hybrid attains AUC $= 0.869$ ($\Delta\mathrm{AUC} = +0.015$ vs.\ LR, $p<0.001$; $-0.004$ vs.\ XGBoost), cuts Brier from $0.158$ to $0.085$ (a 46\% reduction), and replicates the regional routing pattern. The fairness audit detects severe routing into the opaque ML-driven region along socioeconomic axes: rural respondents by 18 percentage points relative to urban, primary-or-less-educated by 32 points relative to secondary-and-above, and Ugandan respondents by 22 points relative to Kenyan, while gender shows essentially no routing disparity. The audit pipeline surfaces subgroup-routing violations that aggregate fairness metrics miss, in a form directly usable by African central-bank supervisors of digital credit.

Explainable AI-Based Credit Default Risk Modeling on Multi-Table Data

This study addresses the prediction of credit default risk using the Home Credit Default Risk dataset. Due to the approximately eight percent imbalanced class distribution in the data structure, an intensive feature engineering pipeline that reduces multiple tables to the customer level was applied before modeling. The implementation was designed to include behavioral summaries generated from external credit history, previous applications, installment behavior, credit card transactions, and POS cash records. Missing value handling, encoding, and multicollinearity reduction steps were performed on the obtained features. LightGBM was selected as the classification model, and the hyperparameters were optimized with Optuna under cross validation. In the best configuration, the mean AUC value was observed as 0.78682 and the mean PR AUC value as 0.28015. In order for the probabilities to be used more reliably in decision making processes, Platt scaling and SHAP based explainability analysis were applied.

Dependent latent class partial credit models for careless and insufficient effort responding in online survey data

$\texttt{findr}$: Transparent and Fair Credit Risk Decisions through Semi-Structured Regressions

Credit risk models increasingly need to combine predictive accuracy with transparent explanations and auditable fairness constraints. Logistic regression remains attractive because its coefficients are easy to interpret, but it can miss nonlinear structure. Flexible models can improve prediction, but their explanations are often post-hoc and may not describe the decision rule itself. We introduce $\texttt{findr}$, short for flexible, interpretable deep regression, a semi-structured framework for binary credit risk modelling that decomposes the logit into an interpretable structured component and an orthogonal neural residual. The orthogonalisation separates coefficient-based effects from residual nonlinear variation, while an in-processing Wasserstein penalty mitigates group disparities by comparing score distributions during training. The framework also includes diagnostics that measure the structured component's contribution to logit variation, decision agreement, and local directional consistency. We evaluate $\texttt{findr}$ in a simulation study and on eight public credit datasets using score-level accuracy-fairness frontiers. The results show that $\texttt{findr}$ behaves close to logistic regression when the signal is approximately linear, while recovering much of the predictive gain of neural models when nonlinear structure is relevant. The diagnostics identify when coefficient-based explanations remain close to the full fitted model and when residual variation must also be examined. These findings support semi-structured modelling as a practical way to make performance, fairness, and interpretability trade-offs explicit in credit risk decisions.

The Effect of Data Interventions on Local Explanations in Credit Risk Models