Default Probability Prediction by Explainable Machine Learning for Small and Medium-Sized Enterprises

Shadpour, Mehdi

4. semester

Economics and Business Administration (Finance), MSc.

2024

2024-06-02

55

In this paper it is attempted to assess the credit risk of small and medium-sized enterprises (SMEs) via explainable machine learning through different machine learning models. Although SMEs are vital for economic growth and so lending loan to them, their credit worthiness evaluation is a burdensome and complex task. Nowadays, Artificial Intelligence (AI) is vastly used by finance sector to provide accurate models especially for credit risk of loan borrowers. By the same trend in this research, the three groups of SMEs in USA which have received the aid program during Covid pandemic (Payroll Protection Program) were examined for forecasting their default probability. Both machine learning algorithm models and explainable AI (XAI) were hired to construct a precise model for predicting each SME’s default probability. Despite returning accurate result by using machine learning models such as eXtreme Gradient Boost (XGB), Logistic Regression (LR), Support Vector Machine (SVM), these models cannot explain the importance of each feature in the default prediction decision. However, by using XAI such as Shap and LIME, besides more accurate results, the importance and effect of each parameter is observable and can be employed to interpret the model’s decision. The research findings proved that XAI models are more accurate, transparent and comprehensive which could assist the financial decision makers to efficiently predict SMEs default probability.

Credit Risk ; SMEs ; Default Probability Prediction ; Machine Learning ; Explainable AI ; Loan Status ; XAI ; Machine Learning Algorithm ; Feature Importance ; Financial Risk ; Loan Lender ; SMEs during Covid Pandemic

Download
View record in AAU Student Projects