A Review And Improvement Of Empirical Research Of Financial Distress Of Listed Companies

This paper firstly reviews financial distress researches home and broad comprehensively and concludes that financial distress study mainly consists of normative theories discussing the cause of distress, the empirical researches for financial distress prediction, and the outcome caused by financial distress. Among the three aspects, empirical distress prediction plays a leading role. Hence, this paper reviews this part of research in detailed classification and makes further improvement in terms of sample interval, prediction model evaluation and the definition of financial distress in its empirical analysis.On the first two points, this paper samples all ST firms and matched non-ST firms from 1998 through 2003 (sample capacity 424 firms) and sets up multivariate linear discriminant models and logistic regression models across 6 years. Compared with other domestic researches, the result of this paper shows that the prediction model based on (t-2) year data achieves a higher accuracy rate of 87.24% than that of other models based on (t-1) year data. Therefore , the linear discriminant model and logistic regression model has a stronger prediction power as early as 2 years prior to financial distress.On the thud point of empirical improvement, this paper evaluates the power of linear discriminant model prediction under broadened definition of financial distress. This paper samples financial distress firms (including ST firms and hidden fiancial distress firms) in 2002 and matched sample and builds up a multivariate linear discriminant model adopting 15 commonly-used variables. The model achieves an accuracy rate of 91.4%. In order to verify the prediction ability of the model, this paper uses validation sample in 2003 and still achieves an accuracy rate of 82.7%. This paper finally analyses the empirical results and puts forward its views on the future application and development of financial distress research in China.