Corporate Financial Distress Predicting with Machine Learning Techniques

Abstract

This dissertation presents a study into the utilization of machine learning models for Corporate Financial Distress Prediction of small and medium-sized enterprises (SMEs) in Italy. A comprehensive examination of the relevant literature is conducted, including an overview of traditional financial distress prediction methods, such as Logistic Regression, as well as the utilization of machine learning in financial distress forecasting. The study aims to evaluate the efficacy of various machine learning techniques, namely Random Forests and Neural Networks, in predicting corporate failure by constructing and training predictive models using financial indicator data from a sample of SMEs.