Logistic Regression Analysis for Prediction of Financial Failure: Evidence from Central Public Sector Enterprises in India

Abstract

The present study is intended to predict the financial failure of Central Public Sector Enterprises (CPSEs) in India using financial factors that cause the failure and show how the probability of failure can be effectively explained. This study is obvious because of the growing failure of the enterprises in India and the factors that push them to fail obviously calls into question the sustainable financial health of these enterprises. Policies, regulations and new strategies should be developed to help management and policymakers to examine the factors that affect the likelihood of failure. For this study, 27 heavy, medium and light engineering enterprises were chosen as a sample, with a 10-year study period. The magnitude of firm-specific endogenous factors in determining and/or explaining the failure of enterprises is revealed by principal component analysis. Binary logistic regression was used to examine these variables. The result of logistic regression has an accuracy of 83.9% in predicting the failure. According to the findings, working capital, net profit, return on assets, gross value added to capital employed, labour cost to sales, capital–output ratio and sales to total assets are the financial factors that significantly impact the probability of failure. Financial health was also examined using the Altman Z-score model. The results demonstrate the negative Z-score recorded by failure enterprises and distressed category enterprises. The study shows that the CPSEs failure can be avoided if indications and influencing factors are established in time and the correct prediction model is applied to enhance the financial situation.