Research On Financial Early Warning Statistics And Intelligent Modeling Of Listed Companies

With the rapid development of society and the continuous improvement of the economic level,the emerging competitive methods make it especially difficult for enterprises to survive and develop.For enterprises which survive,they should always pay attention to the sustainable development of enterprises.As is known to all,there are many factors that affect the business performance of enterprises,which can be viewed from both macro and micro levels.Macro factors refers to the big international economic situation and national policy guidance,which requires the enterprise operator ease pattern,largely to the macroeconomic situation has a relatively correct grasp,adept,moderate to adjust enterprise development strategy,keep up with the economic situation and the trend.And microscopic factors mainly from the enterprise internal,strict internal control system,discover the enterprise internal risk,from two aspects of the management at the grass-roots level,to enterprise's production and circulation have a accurate control.Enterprise external risk we can't control,but for internal business risks,through analysis of enterprise financial status of the previous year to conduct a preliminary judgment,can through the establishment of the corresponding financial early warning model to determine whether there is a risk.This warning is especially important for listed companies,because their financial situation is not only related to the survival and development of the enterprise itself,but also to whether investors can make the right investment decisions.This has important implications for creditors,investors and managers.This article used different methods to research and analyze the financial status of listed companies from the perspective of statistics and intelligent modeling,in the hope to establish a reasonable early warning model to predict the possibility of financial crisis of listed companies.In this paper,Models are established based on Lasso-Logistic regression?KVM model and support vector machine(SVM)? generalized regression neural network,and the market as a whole based on factor analysis and Logistic regression method of predictive model,respectively from four different angles,and reflect the financial position of the enterprise for some time,through the model test,inspection,found that the accuracy is higher.They have improved the common Logistic model in the prediction of timeliness,sample size and statistical and intelligent modeling technology,and achieved a better early warning effect.At the end of the paper,four situations are summarized in order to analyze the financial situation of listed companies in different ways,and the decision makers can make a relatively comprehensive consideration.Finally,it points out the shortcomings of the financial warning and the prospect of development.