A Comparative Analysis of Machine Learning Classifiers and Ensemble Techniques in Financial Distress Prediction

A company with financial difficulties is referred to as Financially Distressed. The aim of this study is to analyze and compare the machine learning classifiers and ensemble techniques in Financial Distress Prediction. Initial research works in this field use intelligent and linear methods for building predictive models. Most of the research works suggest that data mining methods predict financial distress better than traditional methods. This paper aims at building and evaluating machine learning models including Neural Network (NN), Decision tree (DT), and Support Vector Machine (SVM) for Financial Distress Prediction. This paper also focuses on building a more accurate Prediction model using Ensemble techniques including Majority Voting (MV), Random Forest and AdaBoost ensemble, by combining the outputs of individual classifiers. The machine learning models are built using the dataset containing financial data from UAE firms in the period of 2010-2017.