Research On The Financial Crisis Early Warning Model Based On Neural Network

Along with the China's securities market's development and the intense competition day after day, the listed companies become the benefit focal point by all quarters. The reasons is the business efficiency of many listed companies is low, lead the finance condition to worsen suddenly,steped to the bankrupt edge. How to perceive it and inform the related benefit main body before the financial crisis is arrive, becomes the question need to soluted immediately. This article take the neural network method to constructs the financial crisis warning model ,which is based on the manufacturing industry.First, according to the research background and the research significance, this article analyzed and summarized the present situation of the research, summarized the reasons which the financial crisis is produced, and pointed out the significance of the research to the main body related benefit.Next, according to the selection principle of the financial early warning target, by which selected 21 targets which reflect the credit capacity, profit ability and so on, take the statistical method to carry on the parametric test and the nonparameter test with the targets which had selected , finally this article had determined 15 targets,which was used to the independent variables of the model.Then, introduced the application principle of the neural network method in the model.Analysis the characteristic and the domain of the neural network; describe the application which is carries on the model and the unit process of the neural network.Finally, introduced the selected sample, used the BP neural network method, constructed financial crisis forewarn model with the manufacturing industry, take examination and carry on the model in application, summarized the questions and the model limitation which should be paid attention in the application.