The Research Of Financial Distress Prediction On Chinese Manufacturing Listed Companies Of A-share

In 2014, Chinese manufacturing industry is facing a more severe situation. As a result of shortage of labor supply,increase of labor cost and other factors, many manufacturing companies’ financial situation has a big problem. Many manufacturing companies are in financial distress and even go bankruptcy. Therefore, It’s very meaningful to make use of the financial data of manufacturing listed companies and develop a model to find out financial issues to avoid risk.First of all, this paper introduces the theory of financial distress and financial distress prediction, and combsdefinition of the concept of financial distress, causes of financial distress, the meaning and function of financial distress prediction in detail. Theory is introduced to guide empiricalresearsh; this paper will define the corporation that is treated specially by the Commission due to financial problem as the company in financial distress. In order to provide a comprehensive and real prediction of financial distress for Chinese manufacturing listed companies of A-share, this paper selected 171 manufacturing companies as the research sample, including 45 ST companies result from financial issue, as well as 126 matched normal financial companies.Then the research sample is divided into two parts, one part of the 116 companies was used to develop the model, while the remaining 55 companies are as testd samples for testing the model. Based on previous scholars’ analysis on the factors of financial distress, combined with the characteristics of the manufacturing industry, this paper selected 55 indicators, including the 48 financial indicators and 7 non-financial indicators. These indicators reflect the enterprise’s debt paying ability, operating ability, cash flow ability, growth ability, risk, governance structure, ownership structure, audit opinion compresensively. Then this paper test the significant difference of the 55 indicators. Through the T test and non-parametric test to exclude the 25 predictor variables that are not significantly affected, this paper develops a comprehensive prediction index system. In order to eliminate the collinearity between predictors, this paper uses the most principal component analysis, 11 common factors are extracted. The accumulative variance of the 11 factor contribution rate reached 81.552%, containing most of the original information prediction index, so factor extraction is ideal. Finally, this paper uses the 11 factors as explanatory variables to develop a Logistic regression model.The research result show that the prediction of Logistic is 90.5%,predicted rate of the tested sample reached 81.8%, which is a satisfactory prediction result. Therefore we can come to an conclusion that the accuracy rate of the model is very high. It can be applied to the Chinese manufacturing listed companies of A-share, which has practical value.