SHAP-powered machine learning for financial distress prediction in Malaysian firms

Key Points

• The aim is to develop a machine learning model using SHAP to predict financial distress in firms.
• Utilized SHAP (SHapley Additive exPlanations) for model interpretation and prediction analysis.
• Applied various machine learning algorithms to financial data from Malaysian firms.
• Analyzed model performance using relevant financial metrics.
• Achieved significant accuracy in predicting financial distress.
• Identified key financial indicators that contribute to distress predictions.
• Validated the model's predictions through cross-validation techniques.

Abstract