Font Size: a A A

Research On Credit Evaluation Of Internet Listed Companies Based On Support Vector Machine

Posted on:2020-09-15Degree:MasterType:Thesis
Country:ChinaCandidate:S TianFull Text:PDF
GTID:2417330578457367Subject:Applied Statistics
Abstract/Summary:PDF Full Text Request
Since the Internet officially entered China on April 20,1994,it has only been more than 20 years,but it has become the underlying structure and standard of the whole society and economy.It has not only subverted the media industry,but also has a revolutionary impact on all walks of life.It has played a positive role in promoting the rise and prosperity of China's economy and is bringing profound impact and incredible changes to human society.With the rapid development of the Internet industry,Chinese Internet enterprises have also developed rapidly in recent years.Internet giants and unicorns have been in the forefront of the world in terms of scale and quantity with American enterprises.Generally speaking,China's Internet economy is highly active and volatile.Services and applications change at a fast pace,the phenomenon of outlets in the Internet industry is more obvious,the number of enterprises in peak period is more,the average life of enterprises is shorter,accordingly,it is easier to create overnight fame.Therefore,the rapid development of the Internet economy has also brought some social credit problems that can not be ignored,such as the problem of dishonest running of the Internet financial platform is often common.How to evaluate the credit of Internet listed companies impartially and independently is not only conducive to the company's more convenient access to investment and financing in the capital market,but also conducive to increasing investor confidence in the company and also conducive to the healthy development of the Internet industry in the future.With the fading of entrepreneurship boom,the number of Chinese Internet start-ups has declined since 2015 after experiencing explosive growth.The Internet market tends to be rational,and the market competition will become more intense.The good credit status of Internet companies is a strong guarantee for the fierce competition in the Internet industry.Credit evaluation will become more and more important in the future development of China's Internet industry,which urgently needs scientific and reasonable credit rating evaluation.Firstly this paper introduces the concepts of Internet and credit evaluation,then summarizes the commonly used credit evaluation methods,then analyses the current situation of Internet companies,and points out the importance of credit evaluation to the development of Internet companies.What' more,it constructs a credit evaluation index system which conforms to the characteristics of Internet listed companies,including six first-level indicators and 16 second-level indicators.It introduces the value of social contribution per share and measures the integrity of the company more scientifically and reasonably.Secondly,we use the support vector machine to model the index system,analyze and compare the classification accuracy under different kernels,and determine the appropriate kernels and penalty factors.The classification accuracy of credit rating of 115 Internet listed companies reaches 92.1739%.At the same time,compared with BP neural network model,it proves that support vector machine has better classification effect in the study of credit evaluation of Internet listed companies.Finally,based on the established support vector machine model with radial basis function as the kernel function,128 Internet listed companies are evaluated for credit,and the current credit situation of Internet listed companies is summarized as follows.At present,97.66%of the Internet listed companies are in credit grade A or above,with high reputation and good development prospects.The overall investment environment of the Internet industry is relatively stable and the investment risk is relatively small,but we still need to pay close attention to the changes of the company's credit status.
Keywords/Search Tags:Internet listed company, credit evaluation, Support Vector Machine, BP Neural Network
PDF Full Text Request
Related items