AI-Based Bankruptcy Prediction for Strategic Decision-Making in Emerging Market Firms

This study investigates the use of artificial intelligence (AI) techniques to predict bankruptcy among emerging market firms, using a structured, synthetic dataset modelled on the financial characteristics of 100 Indian companies across multiple sectors. The dataset includes revenue, profitability, leverage, liquidity, solvency, and cash flow indicators, from which bankruptcy labels were generated using rule-based financial thresholds. Three machine-learning models—Random Forest, Logistic Regression, and XGBoost—were trained and evaluated. Random Forest achieved the highest accuracy and produced the most stable predictions. Feature importance analysis shows Cash Flow Ratio, Interest Coverage, and Net Profit as the strongest predictors of financial distress. Sector-wise results indicate that capital-intensive industries exhibit higher bankruptcy risk. While the results validate AI’s potential for early warning systems in emerging markets, the use of synthetic data limits external validity. The study recommends testing the models on actual Indian financial data for improved generalisation and incorporating additional ratios and cross-validation strategies for enhanced robustness. The findings contribute to business strategy and risk management by demonstrating how AI-driven models can support early assessment of firm vulnerability.

Bankruptcy Prediction, Machine Learning

1. Altman, E.I. (1968). Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. Journal of Finance. [Google Scholar] [Crossref]

2. Ohlson, J.A. (1980). Financial Ratios and Probabilistic Prediction of Bankruptcy. Journal of Accounting Research. [Google Scholar] [Crossref]

3. Barboza, F., Kimura, H., & Altman, E. (2017). Machine learning models and bankruptcy prediction. Expert Systems with Applications. [Google Scholar] [Crossref]

4. Kumar, S., & Rao, P. (2016). Financial distress prediction in emerging markets. International Review of Economics and Finance. [Google Scholar] [Crossref]

5. Zhang, Y., & Wang, S. (2020). XGBoost-based financial distress prediction. Applied Soft Computing. [Google Scholar] [Crossref]

6. Singh, A. (2022). Credit risk assessment models in India: A machine learning approach. Indian Journal of Finance. [Google Scholar] [Crossref]

7. Sharma, A., & Kumar, D. (2022). Machine learning techniques for financial distress prediction in emerging markets: A comparative study. Journal of Risk and Financial Management, 15(4), 145–160. [Google Scholar] [Crossref]

8. Barboza, F., Kimura, H., & Altman, E. I. (2017). Machine learning models and bankruptcy prediction. Expert Systems with Applications, 83, 405–417. [Google Scholar] [Crossref]

9. Liang, D., Tsai, C. F., & Wu, H. T. (2020). The effect of feature selection on bankruptcy prediction accuracy: Empirical comparison of machine-learning techniques. Knowledge-Based Systems, 195, 105– 724. [Google Scholar] [Crossref]

• The Indirect Effect of Liquidity and Activity on Company Value with Profitability as an Intervening Variable
• Effect of Financial Skills, Knowledge, and Attitude on The Financial Behaviour of Clergy
• A Decade of Review: Trends in Budget Execution and Financial Performance of Development Projects in Tanzania (2014/15-2023/24)
• The Influence of Pre-Project Planning on the Budget Absorption Rate of Public Funded Infrastructure Projects in Kenya a Comparative Case Study of Narok, Migori, and Kisii County Government Projects
• Assessment of Factors Influencing Digital Transformation in Hotels’ Facility Management in Abuja Metropolis, Nigeria