Study On Sales Forecasting Model In Commercial Enterprises Based On Time Series Analysis

With the development of information technology, it has accumulated a large numberof historical sales data in management information system of business enterprises in ourcountry. These data are just stored simply in the database of information system, and havenot gotten enough attention. Under the condition of advanced information technology, weshould use the business enterprise’s sales data to establish effectively mathematical modelto realize sales forecasting for commodities. It can assist enterprise decision-makers makeaccurate sales strategy and reduce cost of sales and increase profits for commercialenterprises. It has a certain realistic guiding significance for improving the scientificity andrationality of business enterprise management. This paper adopts the normative studymethod, statistical method, regression analysis, method of combining qualitative analysiswith quantitative analysis, systems analysis and design and software development methods.The main work is as follows:Firstly, the paper summarized the research status inside and outside on issues related.It pointed out the research of combining numerous sales forecast theory and method withthe actual to solve the concrete problem still need be continuing and we need furtherstrengthen the application of the theory and method of sales forecasting in practice. Basedon this, the paper made comprehensive overview on basic theories and methods related,including the concept and function of sales forecasting, the principle and procedure, thetype and method, the precision and cost, and the concept of time series analysis, the typicaltime series analysis models and prediction accuracy of time series analysis model, thecontent made theory foreshadowing for putting forward sales and building forecastingmodel.Then, the paper summarized the present situation of the application of businessenterprise sales model and the existing limitations, and analyzed the present situation andproblems and causes sales of the business enterprise management, and pointed out thenecessity of sales forecasting in business enterprise application. As a foothold, the paperdivided the various factors affecting commercial enterprise sales data into four categories:the season change, trend change, and cycle change and irregular change. Time seriesdecomposition forecasting model is put forward. After pointing out the deficiency existing in this model, the paper made some innovative improvements in the model. Next, the paperapplied the quarter sales data from management information system in QianYe glassesenterprise to time series decomposition forecasting model, in order to verify the feasibilityof the model. On the basis of analyzing the sales data, this paper put forward the idea ofsetting sales index for commercial enterprises, which is one of the paper’s innovationpoints.In the last part of the paper, it analyzed, designed, and developed Business EnterpriseSales Forecasting System, using PowerBuilder9.0as the front-end development tool andSQL Server2000as the backstage database, to put the theoretical results into practice, toassist business enterprise to make scientific and rational management decisions.