Listed Companies' Financial Crisis Early Warning Based On A Different Cycle Studies

A global financial crisis swept the whole world at the end of 2008.The world's various economies had suffered a degree of strike. In the meantime many companies were facing a cash shortage or even bankruptcy. The paper takes this phenomenon as a entry point, the payment crisis in the enterprise financial crisis may occur at any time as a short-term early warning research object, and loss crisis(consecutive losses) as long-term early warning of object, these two parts constitute the main article.The first part in this paper analyze features of the financial crisis and its manifestations deeply, definite the solvency crisis according to this definition, the 54 manufacturing listed companies at the Shanghai Stock Exchange were selected as failed samples, according to the principle of proportion 1:1 in the same industry and same size,54 normal company are matched with them. On the process of forecast variables selection, the cost of financial distress is introduced because the feature of which is more advance information and combined with the previous year's cash flow ratios, through factor analysis, the representative indicators are extracted as input variables to establish Logistic regression models. This model has higher prediction accuracy. The study finds indicators which have prediction function to the shortage of the company's cash flow.In the second part, the research object is the loss crisis, the paper show that compare with payment crisis in the long term, the loss of a crisis is predictable. Therefore this paper takes the ST Listed manufacturing companies as loss crisis samples which loss in 2007 and 2008 for the first time. Using the total of three years'financial data which from ST-3 year to ST-1 year with the same industry, same size, same period, non-ST-listed financial data as the match samples to do a long term research. And also this part study whether the cash flow information have incremental information content to the traditional accounting-type indicators, which takes two steps:in first step,the traditional accounting-type index is used only to Logistic regression model,the second step is add cash flow information to the traditional accounting-type index and a new model is established,the study show that the cash flow information in the ST-2, ST-3 years can effectively improve the prediction accuracy of the model, indicating cash flow information for a loss crisis also has a high predictive value, in the meantime the direct signal is given in early period.Finally, the paper analyzes the results above and as the conclusion of article which provide the basis and inspiration to enterprise financial crisis management..