Application of Machine Learning Tools in Bankruptcy Prediction: A Comparative Study Between Extra Trees Classifier and Logistic Regression

The topic of bankruptcy prediction has garnered significant attention across various disciplines. This paper focuses on the comparison of the accuracy of two distinct prediction models. The methodology employed involves the construction of both an Extra Trees Classifier model and a Logistic Regression model, utilizing a dataset comprising information from 300 Moroccan Small and Medium-sized Enterprises (SMEs). The results of our analysis reveal that the Extra Trees Classifier model exhibits higher accuracy compared to the Logistic Regression model. This comparative study contributes valuable insights to the ongoing efforts in refining bankruptcy prediction models and aids in the selection of appropriate tools for more effective financial risk management.