The Journal of Finance · 2021 · 445 citations · 45 references
BusinessEconomicsFinanceFinancial EconomicsMachine LearningMicroeconomicsCredit MarketsCredit RiskCredit UnderwritingFairness In Machine LearningMortgage LendingMortgage PolicyAlgorithmic Fairness In UnderwritingCredit Risk Machine Learning ModelsDefault Risk PredictionEquilibriumLoan Default Prediction
Innovations in statistical technology for credit screening raise concerns about racial distributional impacts, as greater flexibility may uncover structural relationships or triangulate excluded characteristics. The study predicts mortgage default using both traditional statistical models and machine learning models on U.S. mortgage data.
ABSTRACT Innovations in statistical technology in functions including credit‐screening have raised concerns about distributional impacts across categories such as race. Theoretically, distributional effects of better statistical technology can come from greater flexibility to uncover structural relationships or from triangulation of otherwise excluded characteristics. Using data on U.S. mortgages, we predict default using traditional and machine learning models. We find that Black and Hispanic borrowers are disproportionately less likely to gain from the introduction of machine learning. In a simple equilibrium credit market model, machine learning increases disparity in rates between and within groups, with these changes attributable primarily to greater flexibility.
45
Leo Breiman · Machine Learning · 2001 · 126.2K citations · Full text
The relationship between Precision-Recall and ROC curves
Jesse Davis, Mark Goadrich · 2006 · 6.3K citations · Full text