Enterprise financial distress prediction based on BPNN: A case study of Chinese listed companies

Enterprise financial distress prediction has been the attention focus in the theory study and the business community. To build a scientific, fast and effective model for financial crisis prediction of the Chinese listed companies, 11 key financial indicators are chosen for the financial distress prediction model. The factor analysis is used to extract five common factors and therefore, to get the comprehensive score of each sample. The traditional ST and non-ST classification criteria are abandoned; the score intervals of the enterprise financial status are divided in a novel way-health, concern and distress. Finally, the five common factors are trained and tested as the input and the financial status as the output with the prediction model based on the backpropagation neural network. The result shows that the proposed model is accurate and can provide a great assistance for enterprises, investors and decision-makers.