Research On Financial Early Warning Model Of Manufacturing Industry Listing Corporation In China Based On Support Vector Machine

With the rapid development of China's market economy and the continuous improvement of the capital market, listing Corporation in China continue to grow and develop, however, the market is complex and changeable, opportunities and challenges coexist, the listing Corporation in China are also facing new challenges. Manufacturing industry is the pillar industry of the national economy, China's manufacturing industry accounted for the proportion of the total number of listing Corporation and listing Corporation in 60% above, with a major impact on the national economy. However, in recent years our country manufacturing industry listing Corporation due to financial difficulties and meet the eye everywhere by special treatment, therefore, is of great significance to study how to prevent financial crisis of listed manufacturing companies.This paper first reviews the domestic and foreign scholars in the study of early warning model of financial results, expounds the definition and characteristics of financial crisis, defines the definition, function and the steps of financial early warning, analyzes the reasons of China's manufacturing industry enterprise's financial crisis from different aspects. On the basis of this, this paper selects 26 companies listed in Shanghai and Shenzhen stock market A advocate board in 2013 and 2014 were specific manufacturing processing listing Corporation, according to the principle of similar asset size, select the number of the same financial normal manufacturing listing Corporation as the paired samples, the listing Corporation is the financial data of two years ago as a special treatment during the study period, select the financial data of 36 company as the training samples, the financial data of 16 companies as test samples. In the aspect of financial early warning index system, this paper selects 21 financial indicators to reflect the financial situation of enterprises from different aspects. Finally, based on the support vector machine algorithm, constructed three for manufacturing listing Corporation financial early-warning models, including the main contents:(1) In view of the ability to distinguish between different financial indexes in the financial crisis companies and the non-financial crisis companies of different, presents a significant test of financial early warning model and support vector machine based on index. The model through the significant test on financial index, find out the financial index can best distinguish the financial crisis company and the financial crisis, and then construct the model of support vector machine. The experimental results show that, the model of 16 test samples of all the discriminant accuracy, with discriminant accuracy rate is very high.(2) In view of the fact that for more financial indicators to construct the financial early-warning model, there is a certain correlation between the indexes, directly used to predict the effect modeling effect of model, this paper will be the method of principal component analysis and support vector machine are combined to establish the financial early-warning model. The model first using the method of principal component analysis, principal component analysis is conducted on the financial index, and constructs the model of support vector machine. Empirical analysis shows that, the method of discriminant accuracy rate is 93.75%, which can effectively predict the financial crisis.(3) Whereas the predictive ability of the algorithm of support vector machine is largely influenced by the selection of parameters, put forward a kind of particle swarm algorithm to optimize the financial early warning model based on support vector machine. The optimization of the parameters of support vector machine is made by using the particle swarm algorithm, in order to improve the performance of the algorithm. The experimental results show that, the model forecast accuracy rate is 93.75%, suitable for manufacturing listing Corporation financial crisis prediction.