Predicting Financial Distress with XGBoost: Evidence in the Latin American Public Utility Sector
DOI:
https://doi.org/10.11606/issn.1982-6486.rco.2026.243777Keywords:
Risk management, Credit risk, Machine learning, Emerging markets, InsolvencyAbstract
This study investigates the effectiveness of the XGBoost machine learning algorithm for forecasting financial distress (FD) in Latin American utility firms using financial data from 2000 to 2019. The dataset comprises firms from six Latin American countries, with FD defined through accounting indicators, such as negative Earnings Before Interest, Taxes, Depreciation, and Amortization (EBITDA), and market indicators based on declines in market value. The model was trained using undersampled data to address class imbalance and evaluated using metrics including Area Under the Curve (AUC), accuracy, Type I and Type II errors, and Brier Score. The results show that, within the sector-specific sample analyzed, the XGBoost model outperformed logistic regression and achieved performance comparable to or superior to that of Random Forest, whose reference results were obtained from a multi-sector sample. Overall, XGBoost achieved accuracy ranging from 88% to 95.5% and Type II error rates between 0% and 12%, surpassing the reference model. The most relevant variables differed according to the type of financial distress. For accounting-based FD, profitability indicators, especially earnings per share, were predominant. For market-based FD, variables associated with investor perceptions became more prominent, including year-over-year changes in the price-to-book ratio and location-related factors (country-level variables). The study confirms the potential of XGBoost both for forecasting financial difficulties with a one-year lead time and for interpreting the underlying factors, which evidences a potentially useful tool for analysts, regulators, and investors. More broadly, the results contribute to the ongoing discussion of forecasting performance in finance.
Downloads
References
Altman, E. I. (1968). Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. The Journal of Finance, 23(4), 589–609. https://doi.org/10.2307/2978933
Araújo, M. M. N., Sousa, A. B. V., Parente, P. H. N., & Santos, J. G. C. dos. (2025). Subsídios governamentais e risco de insolvência nas instituições financeiras brasileiras. Revista de Contabilidade e Organizações, 19, e232628. https://doi.org/10.11606/issn.1982-6486.rco.2025.232628
Barboza, F. (2023). Data and Code (in R) for predicting Financial Distress in Latin America (Versão 1) [Conjunto de dados]. Harvard Dataverse. https://doi.org/10.7910/DVN/ODLGNJ
Barboza, F., & Altman, E. (2024). Predicting financial distress in Latin American companies: A comparative analysis of logistic regression and random forest models. The North American Journal of Economics and Finance, 72, 102158. https://doi.org/10.1016/j.najef.2024.102158
Beaver, W. H. (1966). Financial Ratios As Predictors of Failure. Journal of Accounting Research, 4(3), 71–111. https://doi.org/10.2307/2490171
Canton, C., Müller, M., da Silva, T. P., & da Rocha Armada, M. J. (2021). Dificultad financiera y rendimiento de mercado de organizaciones latinoamericanas. Revista Facultad de Ciencias Económicas, 29(1), 11–26. https://doi.org/10.18359/rfce.4450
Cardona, L., Gutiérrez, M., & Agudelo, D. A. (2017). Volatility transmission between US and Latin American stock markets: Testing the decoupling hypothesis. Research in International Business and Finance, 39, 115–127. https://doi.org/10.1016/j.ribaf.2016.07.008
Chen, M.-Y. (2011). Predicting corporate financial distress based on integration of decision tree classification and logistic regression. Expert Systems with Applications, 38(9), 11261–11272. https://doi.org/10.1016/j.eswa.2011.02.173
Chou, J. S., & Pham, N. N. H. (2026). Boosting ensembles and deep vision networks optimized by forensic-based investigation algorithm for financial distress prediction in construction firms. Journal of Big Data, 13(1), 25. https://doi.org/10.1186/s40537-025-01345-w
Eça, J. P. A., Albanez, T., & Valle, M. R. do. (2025). Um panorama da renegociação de dívidas de empresas brasileiras de capital aberto no período 2010-2021. Revista De Contabilidade E Organizações, 19, e218938. https://doi.org/10.11606/issn.1982-6486.rco.2025.218938
Gabrielli, G., Melioli, A., & Bertini, F. (2026). Corporate financial distress prediction: a machine learning approach in the era of big data. Journal of Accounting & Organizational Change, 22(7), 31-65. https://doi.org/10.1108/JAOC-05-2025-0166
Geng, R., Bose, I., & Chen, X. (2015). Prediction of financial distress: An empirical study of listed Chinese companies using data mining. European Journal of Operational Research, 241(1), 236–247. https://doi.org/10.1016/j.ejor.2014.08.016
Gepp, A., & Kumar, K. (2015). Predicting Financial Distress: A Comparison of Survival Analysis and Decision Tree Techniques. Procedia Computer Science, 54, 396–404. https://doi.org/10.1016/j.procs.2015.06.046
Huang, Y.-P., & Yen, M.-F. (2019). A new perspective of performance comparison among machine learning algorithms for financial distress prediction. Applied Soft Computing, 83, 105663. https://doi.org/10.1016/j.asoc.2019.105663
Kristanti, F. T., Salim, D. F., & Ihsan, A. F. (2025). Financial distress prediction in emerging markets by using a machine learning approach for financial sustainability. Discover Sustainability, 6(1), 1296. https://doi.org/10.1007/s43621-025-02199-1
Ohlson, J. A. (1980). Financial Ratios and the Probabilistic Prediction of Bankruptcy. Journal of Accounting Research, 18(1), 109–131. https://doi.org/10.2307/2490395
Pindado, J., Rodrigues, L., & de la Torre, C. (2008). Estimating financial distress likelihood. Journal of Business Research, 61(9), 995–1003. https://doi.org/10.1016/j.jbusres.2007.10.006
Poffo, R. F., Bubeck, S. K. & Hein, N. (2025). Habilidade gerencial e financial distress: uma análise de empresas listadas na B3. Revista de Contabilidade e Organizações, 19:e230114. https://doi.org/10.11606/issn.1982-6486.rco.2026.230114
Vu, K. M., & Gurtoo, A. (2014). Utility Sector Performance Post Reforms: Investigating the South Asian Economies. Poverty & Public Policy, 6(2), 157–175. https://doi.org/10.1002/POP4.68
Yang, H., Li, E., Cai, Y. F., Li, J., & Yuan, G. X. (2021). The extraction of early warning features for predicting financial distress based on XGBoost model and shap framework. International Journal of Financial Engineering, 08(03), 2141004. https://doi.org/10.1142/s2424786321410048
Downloads
Published
Issue
Section
License
Copyright (c) 2026 João Pedro Silva Mendes, Daniel Vitor Tartari Garruti, Flavio Barboza
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The RCO adopts the Free Open Access policy, under the standard Creative Commons agreement (CC BY-NC-ND 4.0). The agreement provides that:
- Submission of text authorizes its publication and implies commitment that the same material is not being submitted to another journal. The original is considered definitive.
- Authors retain the copyright and grant the journal the right of first publication, with the work simultaneously licensed under the Creative Commons Attribution License which allows the sharing of the work with acknowledgment of authorship and initial publication in this journal.
- Authors are authorized to take additional contracts separately, for non-exclusive distribution of the version of the work published in this journal (e.g. publish in an institutional repository or as a book chapter), with necessary recognition of authorship and initial publication in this journal.
- Authors are allowed and encouraged to publish and distribute their work online (e.g. in institutional repositories or on their personal page) before or during the editorial process, as this can generate productive changes as well as increase the impact and citation of published work (See The Effect of Free Access).
- The journal does not pay copyright to the authors of the published texts.
- The journal's copyright holder, except those already agreed in the Free Open Access Agreement (CC BY-NC-ND 4.0), is the Accounting Department of the Faculty of Economics, Administration and Accounting of Ribeirão Preto of the University of São Paulo.
No submission or publication fees are charged.
Up to 4 authors per article are accepted. Exceptionally duly justified cases may be reviewed by the Executive Committee of the RCO. Exceptional cases are considered as: multi-institutional projects; manuscripts resulting from the collaboration of research groups; or involving large teams for evidence collection, construction of primary data, and comparative experiments.
It is recommended that the authorship be ordered by contribution of each of the individuals listed as authors, especially in the design and planning of the research project, in obtaining or analyzing and interpreting data, and writing. Authors must declare the actual contributions of each author, filling the letter to the editor, at the beginning of the submission, taking responsibility for the information given.
Authors are allowed to change throughout the evaluation process and prior to the publication of the manuscript. The Authors should indicate the composition and final order of authorship in the document signed by all those involved when accepted for publication. If the composition and authoring order is different than previously reported in the system, all previously listed authors should be in agreement.
In the case of identification of authorship without merit or contribution (ghost, guest or gift authorship), the RCO follows the procedure recommended by COPE.
