Research On The Application Of Neural Network In The Sales Forecasting System Of Chain Enterprises

In recent years,management accounting plays a more and more important role in enterprise operation,and realizes the great transformation of financial value creation.The prediction function of management accounting,forecast by the macro environment analysis and prediction of internal capacity analysis,prediction of profit and cost analysis,continuous guidance business forward,for the sustainable development of enterprises into a controllable and healthy course.In the era of big data,data information is increasing at a geometric rate,and the data analysis of management accounting is facing new challenges.It is worth considering how to better excavate valuable information in the data,raise the level of enterprise management accounting information,and provide support for management decision.As one of the important contents of management accounting,economic forecast plays an important role in planning enterprise's future,supporting management decision and improving management level.But the chain enterprises are faced with the challenge of consumer behavior change and strong development of e-commerce,how to apply the method of information mining data value,forecast the enterprise sales,help management make the right decision for business investment,earn a space in the increasingly fierce competition in the market,it is particularly important.The application of neural network in the field of sales forecasting has been widely verified and applied by scholars at home and abroad.Therefore,this paper discusses the application of neural network sales forecasting technology in chain enterprises.This article uses the standard research method to carry on the discussion to the chain enterprise sale forecast question,divides into six parts.The first part is the introduction.Starting from the research background,this paper puts forward the theoretical significance and practical value of the research on the sales forecasting of chain enterprises,and makes clear the purpose and content of this study.The second part is literature review and review.It sorts out and summarizes the relevant literature at home and abroad from three aspects of the management accounting information and data mining and analysis technology integration,sales forecasting methods research and application of neural networks.Thethird part is the theoretical basis,expounds the relevant theoretical basis of this article.The fourth part is to build model and network training.Use BP neural network,factors from the two aspects of macro environment and internal factors and XX chain stores sales data in Chongqing during 2012-2015 to construct a sales forecasting model for chain enterprises.The fifth part is the design and development of the system.Using MATLAB's GUI platform,we design and develop a chain store sales forecasting system to make it an effective way to use enterprise information system data,and to improve the level of enterprise management accounting information construction.The sixth part is the conclusion and research prospect.It analyzes the innovation of the theory and practice of the sales forecasting system of chain stores,and points out the areas for further study.