Evaluating Corporate Default Risk with Neural Networks: a case study on Italian trade companies.

The thesis deals with the application of neural network architecture in the field of credit risk as a tool for predicting corporate insolvency. Nowadays, artificial intelligence is present in our lives and has inevitably brought about changes in our approach to everyday life. For this reason, it was decided to apply one of its components through the development of a machine learning model for predicting corporate bankruptcy. The first two chapters introduce the concept of credit risk and the various models used to assess it. The third chapter explores the various neural network architectures, focusing on the one of interest for the development of the model. This is followed by a case study based on a sample of Italian companies operating in the trade sector, with the aim of analysing the predictive capacity of the estimated model.