Accurately forecasting financial distress in companies is crucial in the turbulent economic conditions of our time. This study highlights the potential benefits of incorporating qualitative data into financial distress prediction models. The study assessed the relative effectiveness of traditional distress prediction models against integrated models, determined which variables significantly impacted the predictive performance and ascertained the consistency of the models across Africa.The study employed three distinct classification techniques to evaluate the performance of both models: logistic regression, decision trees, and random forests, to ensure that the best-performing technique was identified. The study found that incorporating governance factors into the model did not positively impact the model's performance, affirming that traditional distress prediction models are relatively effective. The study also found that Current Ratio, ROA, ROE, DCE, and Asset Turnover significantly impacted the predictive performance of the models. Finally, it identified regional discrepancies in the performance of the analyzed models.
Author Keywords: Decision Tree, Financial Distress, Integrated Models, Logistic Regression, Random Forest, Traditional Models