Financial Distress Prediction of BSE-SME firms using Support Vector Machine

Abstract: The aptitude to predict financial distress has become an important research topic of today because it could provide an early warning to the company. The current research mainly focused on the accuracy of the model of Support Vector Machine and Logistic Regression. Also this study identified the variables, which could contribute more to the accuracy of the model. The study used financial ratios on the BSE- listed Small Medium Enterprises, during the period of 2015 to 2019. The study found that the logit model yielded an accuracy rate of 85.55% but the SVM- Polynomial model yielded 97.3% accuracy. The sample financial variables such as IBT on Sales, EBIT/ Total Assets, Operating Cash Flow/ Current Liability Ratio, Growth on Interest Before Tax, Working Capital / Sales Ratio and, Return on Capital Employed have significantly contributed to the prediction of the financial distress of sample Small and Medium Enterprises. The study concluded that the SVM Polynomial model was a more balanced and efficient one. Keywords: Financial distress prediction, BSE-SMEs, Support Vector Machine, Financial Ratios, Logit Analysis. Title: Financial Distress Prediction of BSE-SME firms using Support Vector Machine Author: Philo Daisy Rani L., Selvam M. International Journal of Recent Research in Commerce Economics and Management (IJRRCEM) ISSN 2349-7807 Vol. 12, Issue 3, July 2025 - September 2025 Page No: 21-31 Paper Publications Website: www.paperpublications.org Published Date: 30-August-2025 DOI: https://doi.org/10.5281/zenodo.17008237 Paper Download Link (Source) https://www.paperpublications.org/upload/book/Financial%20Distress%20Prediction-30082025-3.pdf