Research On Performance Evaluation Of Internet Finance Listed Companies Based On BP Neural Network

As China’s Internet finance is still at a rapid development stage,scholars’ research on Internet finance is mostly qualitative analysis,mainly from the concepts and models of Internet finance,the risks and regulatory recommendations of Internet finance,the influence of internet finance,etc.Research,there is only a small amount of literature to quantitatively analyze the performance evaluation of Internet financial companies.44 Internet financial listed companies were selected from the WIND database as the research object,and explores the performance evaluation system of Internet financial companies through index selection and scoring models based on the 2015-2018 data disclosed by listed companies.A total of 14 indicators and non-financial indicators were used to build a performance evaluation index system for listed Internet financial companies;data from 44 listed Internet financial companies were selected,normalized and tested for correlation,and the analytic hierarchy process was used to obtain index weights at all levels To get the expected output of the BP neural network.Among them,152 data from 38 companies in the past four years are used as training samples,and 6 companies’ data from the last four years are used as test samples to analyze the simulation output results.As the result,Oriental Fortune got the best comprehensive performance,followed by Lakala and 2345,and China Science and Technology Financial has the lowest comprehensive performance level.The overall performance levels of several Internet financial companies have formed obvious echelon differences.Among those listed companies,the comprehensive performance results have decreased since 2015,and the decline has been relatively large,generally at a low level.It is worth mentioning that,with the stricter supervision of the Internet financial industry and the decline in overall performance for several consecutive years,2017 has still achieved a certain growth,which shows that the Internet financial industry still has some room for development.Fortune’s comprehensive performance score is far ahead of other companies,and it is worth in-depth analysis to provide ideas for the sustainable development of Internet finance companies.The Internet financial companies’ comprehensive performance can be evaluated by the model within a reasonable error range(average absolute error value is 0.018).By finding a reasonable threshold,and then put the statas into a rained network,it can achieve the comprehensive performance measurement and forecast the future development potential of any Internet financial company,and provide a certain reference for the government to formulate policies.It has good practicability and operability.