Research On Financial Early Warning Of Manufacturing Listed Companies Based On BP Neural Network Model

In recent years, with the rapid development of China economy and the stock market, there is fierce competition between enterprises, the companies will face the risk of financial distress at any time, particularly in the manufacturing enterprises. Therefore, the manufacturing enterprises in the financial crisis early-warning, has very important practical significance.This paper constructs a financial early-warning system of manufacturing listed companies. And the warning analysis was carried out by using the BP neural network model. This paper select175manufacturing listed companies of the same size and the same period, in which,115manufacturing listed companies are as the training sample, and60companies are as test samples.The following three aspects are this thesis’ features:First, this paper is different from the previous studies that only use some empirical indicators by virtue of experience, this paper carried out factors analysis of21factors, and selected8representative financial indicators. The paper reduced many variables to a small number of representative variables, however, the amount of information did not reduce, thus, reducing the complexity of model. Second, BP neural network model has high accuracy in manufacturing listed companies financial early warning, and can be widely applied. Third, this paper use case analysis, further proved that BP neural network model fit for the analysis of the financial early warning of manufacturing listed companies.The conclusions of this paper are as follows:firstly, by selecting the same stage, the similar scale, and the same time of the Manufacturing Listed Companies as the research object, the forecast results are good. Secondly, compared with the single variable model, Z-Score model, and logistic model, BP neural network model has more higher forecasting accuracy, and more applicability. Thirdly, BP neural network model for forecasting the three kinds of financial situation of the Manufacturing Listed Companies in our country, the forecast accuracy can reach86.7%. Finally, this paper builds the Manufacturing Listed Companies financial crisis early warning model that can help investors and creditors to correctly judge the financial situation of these companies, when they make investment and financing decisions.