| Financial distress prediction is always an important topic in finance. Since the 1960s, along with global economic integration and the rapid development of countries'economic, Enterprises, especially its'financial problems have become an increasing concern, because of the financial distress always led to the business in trouble or even bankruptcy. Enterprise itself and its investors, creditors and other stakeholders would like to be informed the business risks and crises in time, at the same time scholars have tried to analysis and forecast the financial distress which enterprises will be faced by. So in the near 50 years the researches of enterprise's financial distress has been the focus of the theorists, from the original single-variable model, multiple-variable analysis, the Logistic regression model, to the neural network model and other nonparametric model there are lots of results of these researches has been released. In China many scholars have studied the prediction of listed companies'financial distress, but mainly of them just learned from foreign researcher's results, and at the same time they almost use the industry-wide data, because of the data and sample size. Therefore, to study the use of the existing foreign research results, combined with Chinese actual business environment, to forecast the financial distress is what we need to do.This research based on Chinese listed companies, use theoretical analysis combined with empirical analysis methods, started with the pharmaceutical industry of Chinese listed companies'financial data, using companies which was special treatment (ST) because of the special financial situation as financial distressed company, select the companies financial index covered by profitability, operational efficiency, etc, eight aspects of 22 financial indicators, then use non-parametric test to test whether the two types of company's financial indicators was significantly different, and use factor analysis to condense the financial indicators, select the indicators which were representative as the model variables, then use Logistic regression, to build Chinese pharmaceutical industry listed company's financial distress prediction model respectively in year (t-1) and year(t-2), finally we use the estimate samples and the test samples'data to test the model's forecast accuracy and the test results showed that the integrated correct rate reached 83.85 percent and 80.12 percent. |