New Approaches to Financial and Bankruptcy Risk

Abstract

A consistent direction in which financial risk and bankruptcy analysis models were developed was the inclusion of artificial intelligence algorithms in the methodology, they are being used in most of the cases to achieve some classifications. The artificial intelligence (machine learning) algorithms widely used for the analysis of financial or bankruptcy risks, presented in the paper, are: KNN (K-Nearest Neighbor) algorithm; Support Vector Machine (SVM); Random Forest; Neural networks (ANN – Artificial Neural Networks). Using these algorithms, companies can be classified into different categories, based on some variables, and the final result is to obtain a certain probability of bankruptcy or insolvency for that company. Obviously, there are limitations of the models and problems that can arise from their estimation, among the most well-known being overfitting (the risk of learning the model to perform very well only for the data series being used on). In recent years, ESG (Environmental, Social and Governance) factors have played a very important role. We believe that this is a direction in which the analysis of bankruptcy risk and financial risks could go, by including sustainability aspects in the models.

Keywords

• artificial intelligence algorithms
• sustainability
• corporate governance