Research On The Value Assessment Method Of Pharmaceutical Manufacturing Enterprises Based On BP Neural Network

As an important part of China’s national economy,the pharmaceutical manufacturing industry is closely related to people’s livelihood,and is a key area of the "Made in China 2025" action plan and strategic emerging industries,as well as an important guarantee for building a healthy China.Pharmaceutical manufacturing enterprises need to invest a lot in developing new medicines or factory building,and equity financing inevitably requires transactions in the market.While clarity of enterprise value is a prerequisite for trading in the market,a variety of factors make traditional value assessment methods limited in assessing pharmaceutical manufacturing companies.While BP neural networks have a wide range of applications in many fields,it has a strong non-linear mapping capability,suitable for solving problems with non-linear models,strong adaptive and self-learning capabilities,better generalization ability,fault tolerance and powerful information processing capabilities,which can batch and accurately process information and predict results.Therefore,exploring the application of machine learning in the field of enterprise value assessment helps to combine artificial intelligence with value assessment,enrich enterprise value assessment methods,optimise enterprise value assessment models and reduce transaction risks.This thesis adopts a combination of model construction and case study,takes pharmaceutical manufacturing enterprises as the research object,combines machine learning theory and industry characteristics,selects indicators from financial and non-financial indicators,and constructs a BP(back-propagation)neural network value assessment model for pharmaceutical manufacturing industry,in order to improve the enterprise value assessment method for pharmaceutical manufacturing industry.The model was developed to improve the enterprise value assessment method in the pharmaceutical manufacturing industry.After constructing the model,the data of the case company were brought in and valued by BP neural network and Economic Value Added(EVA)method,which is one of the traditional valuation methods,to verify the validity of the model.Through self-learning training and testing of 175 sample data of pharmaceutical manufacturing companies,due to the easy use of the neural network valuation model,the predictions tested for the case companies are more accurate and less subjective than traditional valuation methods,indicating that the use of BP neural networks can compensate for the shortcomings of traditional valuation methods and has great utility in the pharmaceutical manufacturing industry.In addition,it is found that the use of financial indicators alone is not sufficient to construct an accurate model in valuation of pharmaceutical companies,so nonfinancial indicators need to be included.This thesis extends the research on the application of neural networks and enriches the enterprise valuation methods in the pharmaceutical manufacturing industry.