| dc.description.abstract | In this paper, the sales forecast made by the head office or general agent is used as an indicator. Therefore, the sales forecast is based on the monthly trend and the past sales history, and then the sales target is set for each month. In the research of automobile field, many scholars predict the overall sales forecast performance of each country, so the accuracy of the total forecast is also an important information for the production, manufacturing and supply chain. The forecast of the annual sales volume of each brand is a way to evaluate the gap with the competitive brands. Factors that influence the accuracy of the prediction depend on the results of different forecasting techniques and methods used, as well as the impact of significant factors on the researcher′s assumptions and decisions.
This research method is based on the potential customer management data of the dealers, namely, the walk-in customer, the call-in customer, the internet customer, the old customer, the old customer, the friends and relatives customer, the service customer, the outside display customer and the development of the nine customer sources are discussed. The research method is to use regression analysis of the data. The results show that the sources of potential customers′ data are four independent variables, i.e., walk-in customers, call-in customers, old customers and relative customers, which are significant. The accuracy of the prediction results is greater than 90%, except that the prediction accuracy of 2020 is lower than 90%, and the influencing factors need more data to verify. The statistical method is used to estimate the future sales volume. The empirical results show that the best regression method is valid for the practical application. The data used is the source of dealer customer management, so it can provide the sales manager with a more accurate method to estimate the future sales target.
Keywords: sales forecast, regression, sales process | en_US |