A significant challenge in credit risk models for underwriting is the presence of bias in model training data. When most credit risk models are built using only applicants who had been funded for credit, such non-random sampling predominantly influenced by credit policymakers and previous loan performances may introduce sampling bias to the models, and thus alter their prediction of default on loan repayment when screening applications from prospective borrowers. In this paper, we propose a novel data augmentation method that aims to identify and pseudo-label parts of the historically declined loan applications to mitigate sampling bias in the training data. We also introduce a new measure to assess the performance from the business perspective, loan application approval rates at various loan default rate levels. Our proposed methods were compared to the original supervised learning model and the traditional sampling issue remedy techniques in the industry. The experiment and early production results from deployed model show that self-training method with calibrated probability as data augmentation selection criteria improved the ability of credit scoring to differentiate default loan applications and, more importantly, can increase loan approval rate up to 8.8\%, while keeping similar default rate comparing to baselines. The results demonstrate practical implications on how future underwriting model development processes should follow.

Machine learning applied to financial transaction records can predict how likely a small business is to repay a loan. For this purpose we compared a traditional scorecard credit risk model against various machine learning models and found that XGBoost with monotonic constraints outperformed scorecard model by 7% in K-S statistic. To deploy such a machine learning model in production for loan application risk scoring it must comply with lending industry regulations that require lenders to provide understandable and specific reasons for credit decisions. Thus we also developed a loan decision explanation technique based on the ideas of WoE and SHAP. Our research was carried out using a historical dataset of tens of thousands of loans and millions of associated financial transactions. The credit risk scoring model based on XGBoost with monotonic constraints and SHAP explanations described in this paper have been deployed by QuickBooks Capital to assess incoming loan applications since July 2019.