Research on financial warning for Chinese listed companies by using panel data model

This paper studies how to establish models for predicting financial distress in China´s listed companies. We firstly select 26 companies with financial distress and 54 matching companies´ panel data as samples, then use panel data model to conduct an empirical study. The research indicates that: (1) The predictability precision is 91.25%, 92.5%, 91.25% and 87.5% for T-1, T-2, T-3 and T-4, respectively, superior to the previous research. (2) The panel data model warning analysis is of stable predictability, which is suitable in constructing midterm and long-term prediction models.