Bankruptcy Prediction using the XGBoost Algorithm and Variable Importance Feature Engineering

• Published in: Computational economics Vol. 61; no. 2; pp. 715 - 741
• Main Authors: Ben Jabeur, Sami, Stef, Nicolae, Carmona, Pedro
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.02.2023; Springer; Springer Nature B.V
• Subjects: Algorithms; Bankruptcy; Behavioral/Experimental Economics; Big Data; Computer Appl. in Social and Behavioral Sciences; Data mining; Decision makers; Discriminant analysis; Economic Theory/Quantitative Economics/Mathematical Methods; Economics; Economics and Finance; Feature selection; Information technology; Machine learning; Math Applications in Computer Science; Operations Research/Decision Theory; Prediction models; Social media; Specification; Corporate failure; XGBoost; Bankruptcy; Machine learning
• ISSN: 0927-7099; 1572-9974
• DOI: 10.1007/s10614-021-10227-1

The emergence of big data, information technology, and social media provides an enormous amount of information about firms’ current financial health. When facing this abundance of data, decision makers must identify the crucial information to build upon an effective and operative prediction model with a high quality of the estimated output. The feature selection technique can be used to select significant variables without lowering the quality of performance classification. In addition, one of the main goals of bankruptcy prediction is to identify the model specification with the strongest explanatory power. Building on this premise, an improved XGBoost algorithm based on feature importance selection (FS-XGBoost) is proposed. FS-XGBoost is compared with seven machine learning algorithms based on three well-known feature selection methods that are frequently used in bankruptcy prediction: stepwise discriminant analysis, stepwise logistic regression, and partial least squares discriminant analysis (PLS-DA). Our experimental results confirm that FS-XGBoost provides more accurate predictions, outperforming traditional feature selection methods.