MétaCan
Menu
Back to cohort

Exploring the Trade-off Between Accuracy and Transparency in Credit Risk Prediction Models

2025· article· W4415469702 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueTheoretical and Natural Science · 2025
Typearticle
Language
FieldBusiness, Management and Accounting
TopicFinancial Distress and Bankruptcy Prediction
Canadian institutionsUniversity of Toronto
Fundersnot available
KeywordsInterpretabilityLogistic regressionRandom forestDecision treeBoosting (machine learning)Credit riskPredictive modellingMetric (unit)DefaultOrdinary least squares

Abstract

fetched live from OpenAlex

This study investigates four widely used models for credit risk prediction, Logistic Regression, Random Forest, XGBoost, and LightGBM, focusing on their ability to detect loan defaults under imbalanced and complex data conditions. Model performance was assessed using confusion matrices and key metrics including recall, precision and F-scores and analyzing the metric that best aligns with the purpose of lending institutions to evaluate a model’s performance. This study analyzes the rationale for moving from Ordinary Least Squares (OLS) regression to logistic regression. The results indicate that while logistic regression provides transparency, it struggles with non-linear relationships and class imbalance. Random Forest, built on decision trees, improves stability but sacrifices interpretability. Two boosting methods, XGBoost and LightGBM, achieve superior predictive ability and efficiency with even lower transparency. Also introduces the evolution of decision trees into ensemble methods such as Random Forest, XGBoost, and LightGBM, and analyzes the structural differences of the decision trees employed within these models. Overall, the findings highlight the trade-off between ability to identify target customer of leading institutions and interpretability in credit risk modeling.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesScience and technology studies
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.656
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0020.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.002
Science and technology studies0.0010.003
Scholarly communication0.0010.004
Open science0.0000.000
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.023
GPT teacher head0.237
Teacher spread0.214 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it