Research On Value Evaluation Of Listed Companies On Science And Technology Innovation Board Based On Stepwise Regression-BP Neural Network Model

Since the opening of the Sci-tech Innovation Board in July 2019,this market has provided direct financing channels for a lot of companies with technological innovation capabilities but insufficient development funds,injecting vitality into the development of the country’s technological innovation and industrial upgrading.However,the relaxation of the access conditions on the Sci-tech Innovation Board and the particularity of the companies have made traditional valuation methods subject to various restrictions when valuing companies,which has brought challenges to the valuation of companies in this market.Therefore,how to make a reasonable valuation of the companies on the Sci-Tech Innovation Board is a major issue that needs to be solved urgently.In this context,this article discusses the applicability of Stepwise Regression-BP Neural Network in the value evaluation of listed companies on the Science and Technology Innovation Board.First of all,after comparing different value evaluation methods,this article preliminarily proves from the theoretical level that the Stepwise Regression-BP Neural Network method is more suitable for using in the value evaluation of Sci-tech companies.Secondly,this article analyzes the essence and basic principle of the Stepwise Regression-BP Neural Network method and the formation mechanism of enterprise value under this method,and builds a valuation model.According to this model,this article selects the factors that have an important impact on the enterprise value,and uses the stepwise regression model to test the correlation between the independent variable and the dependent variable,and eliminates the variables that cause multicollinearity,and then imports the optimal independent variable into the BP neural network model as input variables to calculate the enterprise value of the evaluation object.The comparison between the output value of the model and the actual value of the enterprise shows that the Stepwise Regression-BP Neural Network Model has good applicability in the value evaluation of listed companies on the Sci-tech Innovation Board.Finally,based on the research conclusions,the article puts forward suggestions that can help improving the accuracy and rationality of the valuation of companies listed on the Sci-tech Innovation Board.This article uses the stepwise regression model to optimize the traditional BP Neural Network model,and analyzes the principle of the new model,which provides basis for the extensive application of Stepwise Regression-BP Neural Network Model.