本文采用了实证分析方法,从我国上市公司中选择了70家制造企业作为样本(包括35家ST企业和35家非ST企业),选定了5个基础分析指标。随机选取50家配对样本,运用logistic回归模型建立财务预警模型,对其余20家检验样本预测准确达到65%以上,这说明logistic模型在预测公司发生财务困境(亏损)方面是比较有效的。 This paper adopts the empirical analysis method, selects 70 manufacturing enterprises from the listed companies in our country as samples (including 35 ST companies and 35 non-ST enterprises), and selects 5 basic analysis indexes. Randomly selected 50 pairs of samples, the use of logistic regression model to establish the financial early warning model, the remaining 20 test samples accurately predicted more than 65%, indicating that the logistic model in the company’s financial distress (loss) is more effective.