The role of associated risk in predicting financial distress: A case study of listed agricultural companies in China

Abstract

This study investigates the predictive capacity of associated risk for financial distress among listed agricultural companies in China. Seven models, including statistical, machine learning, and ensemble methods, are used to evaluate the contribution of associated risk information. Our findings show that incorporating associated risk significantly enhances model performance, reducing misclassification rates by 0.1 %-3.1 % for healthy companies and 10.8 %-40.6 % for distressed companies, with Random Forest achieving the highest accuracy (0.9523). By incorporating associated risk, the ability of models to identify financially distressed companies is improved. Effective risk identification reduces the accumulation and outbreak of systemic financial risks, providing valuable insights for banking regulatory agencies.