Anticipation of default: implementation of a credit scoring model and integration with macroeconomic factors

This thesis explores the anticipation of credit default through the development and implementation of a credit model. Anchored in the context of the Italian banking sector, it reviews credit risk regulations, such as Basel III and IFRS 9, and their implications for risk management strategies. Leveraging advanced machine learning techniques, including gradient boosting and random forests, the study aims to design an early warning system that predicts defaults with greater precision. Furthermore, the model incorporates key macroeconomic variables, such as inflation, house prices, and interest rate fluctuations, to assess their impact on credit risk. The purpose of the study is to determine whether advanced ML techniques can significantly improve the credit risk monitoring process by intercepting the majority of positions that exhibit credit difficulties and whether the inclusion of macroeconomic features leads to different results.