Research On Credit Risk Of Internet Financial Companies Based On Principal Component Analysis

With the development of e-commerce,big data,cloud computing and other technologies,the functions of the Internet have been continuously broadened.New economic modes such as Internet finance have emerged.Since the "first year" of Internet Finance in 2013,the Internet finance industry has been expanding rapidly because of the explosive development of the Internet in China and the popularity of electronic terminal equipment and the release of the long suppressed investment demand.However,the Internet finance combines financial attributes and Internet attributes,as well as the integration of traditional financial risks and modern Internet risks.The risk is more diversified and complicated,which poses a greater challenge to China's financial supervision.How to prevent credit risk has become the top priority of the next development of Internet financial industry.In 2019,the government increased the supervision of Internet finance.The PBC also proposed to basically solve the risk problem of Internet Finance in 2020.So,this paper studies the credit risk of Internet Finance in China.In this paper,we read the relevant literature of Internet Finance and related credit risk measurement at home and abroad.We combine the development background and development stage and main operation mode of Internet Finance in China.So,we construct a set of systematic Internet financial credit evaluation and measurement system based on the financial statement data of related Internet Finance listed companies.Through the analysis of several common credit risk measurement methods at home and abroad,we find the logistic regression model is more suitable for China's national conditions.Therefore,this paper selects 63 companies from 139 listed Internet financial companies in Shanghai and Shenzhen A-share market from 2015 to 2018,and 252 sets of data are selected as the research samples.Then we select 22 initial indicators from the four-year financial indicators,including profitability,solvency,growth capacity,operating capacity,capital structure and corporate governance factors.Through normal analysis and significance test,the 22 indexes are further screened out.Then the 12 financial indexes are subject to principal component analysis to achieve dimension reduction.Most of the information of the original variables is extracted into 6 factors,and then the 6 factors extracted and retained are used for subsequent regression analysis with default conditions to build logistic regression model and carry out back judgment test.Thus,the credit risk measurement model of Internet Finance in China is established.This model includes six factors which represent the company's profitability,solvency,growth ability,operation ability and corporate governance factors.Each variable of this model has a strong correlation,and the overall goodness of fit is relatively good.Through the back judgment test of the model,it is found that it has 87.31%prediction accuracy in predicting the credit risk of Internet financial listed companies.Therefore,the model can be used to predict whether Internet financial companies have credit risk.