Financial distress prediction with optimal decision trees based on the optimal sampling probability

Financial distress prediction plays an important role in the decision-making process of stock and bond investors, commercial banks and commercial credit adjusters. The effectiveness of financial distress prediction depends on the processing of sample data and the reasonable integration of multiple prediction results. The main contribution of this paper is a novel tree-based ensemble model for financial distress prediction. We obtain multiple balanced samples with different sampling probabilities. The optimal sampling probability is determined by the maximum geometric-mean values, and we construct optimal decision tree models based on the optimal balanced samples with the optimal sampling probability. The model validation is based on a sample of Chinese listed companies. We also validate the effectiveness of the model in different time windows. The empirical results show that the financial distress prediction performance of the proposed model exceeds that of the comparison models in different time windows. This model can contribute toward better credit risk analysis and management.