Abstrakt

Comparison of financial distress prediction model on real estate listed companies

Rong Li, Wang Jiuling

Take real estate listed companies,for example, we studied the internal connection between the companies marked by STand their financial indexesin the three years before the occurrence of financial distress, wemodel the financial crisis of early warning by statistical (Logistic Model) and artificial intelligence (BP neural networks, support vector machine model). Finally we made a comparison to the predict theaccuracy of the classification and the applicability of the three models. Empirical analysis shows that theaccuracy of support vector machine model is the highest, logisticmodel reflected the lowest accuracy, and the BP neural network stood in the middle. Based on the characteristics of the three types of models, wemade some researches ontheir application