Which SME is worth an investment? An explainable machine learning approach

Abstract

Investments in Small and Medium Enterprise (SME) are facilitated by the availability of advanced machine learning (ML) methods, with high computational power and accuracy. However, despite their high accuracy, complex ML models do not provide sufficient explanation and may not be adequate for informed decision-making. In this paper, we propose an explainable AI model that can be used for analyzing SMEs and, in particular, for predicting their expected return, based on their credit risk and expected profitability. Our model is based on Shapley values which allow the predicted values generated by AI models to be interpreted according to the available explanatory variables. To validate our model we have extracted financial performance indicators from the annual balance sheets of 2049 SMEs. As a result of our empirical analysis, we show that the expected return of SMEs can be well predicted and explained by a set of characteristics deduced from their balance sheets. Therefore, their future behavior including risk and return can be predicted.

Keywords: Credit Risk Management, Explainable AI, Expected Return, Financial Technologies

JEL Classification: G11