The Research Of Listed Enterprise Finance Prediction And Valuation

With the fast developmen of our country market economy, marketing competition is gradually vigorous, financial management faces the larger risk and complexity. Enterprises fall into crisis because of financial risk, even bankrupt in recent years. So how to make a efficacious model of finance risk prediction and valuation, this not only is the key point which the listed company pays attention to, but also become urgent need for various benefit correlations. Support Vector Machine (SVM) is a new kind of machine learning methods based on statistic learning theory. It employs the criteria of structural risk minimization. And it's a quadratic programming problem which can make sure that the extreme solution found is the optimal one. So it can use limited information to obtain statistic principles and high generalization, and can also provide a framework for the small samples, nonlinearity and high dimension problems which most traditional learning methods can't solve. Recently, SVM has been successfully applied to text classification, handwritten digit recognition, face detection and face recognition, the fault diagnosis fields.Taking advantage of the modern intelligent technique, Support Vector Machine is introduced in this paper as a tool used in the enterprises financial prediction and valuation fields . SVM offers a new modeling thought and method for researching the enterprise's financial risk,and widens this research area. In the course of studying, the listed enterprises are separted into ST&*ST and not ST&*ST according to their financial condition based on the general way of security market, which is standard and operable. Aiming at the shortcoming of the traditional models'linear assumpation. SVM utilizes its efficiency in nonlinear classification and improves the precision greatly. This paper firstly retrospecta the history of the home and abroad enterprise finance risk prediction and valuation. The SVM method which had been successfully applied to project field can also be used for the studying of enterprise finance risk prediction and valuation. Then, this paper discusses the classifying theory of the SVM. Be aimed at finance risk problem, give the standard of ascertaining and choosing the studying sample and the financial data, not only including the tradition financial data but also including the cash flow data. Be compared with other study in the past, this paper combines the Principal component Analysis and SVM . Simplify the input by Principal component Analysis and classify the enterprise finance by SVM to realize the effect of forecast and valuation.The demonstration is indicated that the Support vector machine method can be accurate abstracting enterprise finacial status characteristic. Under such data condition as small sample, high dimension, big noise and nonlinear relation ship, the thesis analyzes the enterprise status assessment. This new method overcomes the localization of linearity distinguish in Multivariate Discriminant Analysis models and it need less samples to train, the better result contrast against the BP neural network.