A Research On The Implement Of Automobile Sales Forecast Models

Automobile industry is regard as underpinning of country economy, which has a very important role in economic developments. Recently, automobile industry is increasing with the high speed steps and continual extending scales of auto-market. The strong support of new technology and policy to automobile industry and the improving of people's living standard enlarge our country's automobile market and improve the sale. With the enhancements of informational and enrichments of sales data, it is necessary for managers to do forecast sales. Franchisers need to forecast market requirements of automobile sales volume to make the sale planning. Producer need it the same. They make the suitable produce planning to satisfy the demand of the market and get the advantage according the forecast result, and avoid the unnecessary lose of resource. Consequently, how to find a good data mining methods is a vary importance and worthy research task.Now the researches on automobile sales forecast are mainly that improving the forecast models according data analysis, but seldom use the models to build the operational information system. This thesis describes a sales forecast model with automobile sales volume data. The data mining process starts with the character of historical sale data to optimize forecasting models. The paper adopts radial basis function neural network, gray model to establish two different forecast models. Through the empirical analysis, it compares the forecast models' features, and gives their statement and premise of using the models. Finally it uses the software tools to build the automobile sales analysis and forecast system and make it to be a independent information platform for the follow-up test and application.The whole thesis consists of six chapters. Chapter one introduces the background of the research, puts forward the question, summarizes the research content and structure frame of the thesis. Chapter two gives review to theories of backwardness. It includes the theory of decision support systems and data-mining. The deficiencies of those researches and the direction of the paper are pointed out. Chapter three describes the basic theory of radial basis function neural network and gray model, and establishes their models in automobile sales forecast. Chapter four uses these forecast models to do empirical analysis, compares the forecast models' features and gives the premise of using the models. Chapter five give the system analysis and design on the base of the models. It builds the automobile sales analysis and forecast system. Chapter six is the summary and the forecast. It also gives the shortages on this research.