Application Of Combined Forecasting Method In Prediction Of Cigarette Demand In Cloud Production

Since the Reform and Opening-up, China’s tobacco experienced from small to large, from weak to strong, from scattered to centralized, from inferior to major suit development etc. At present, because of domestic and international environmental changes, economic downward pressure on the impact of China’s tobacco industry is facing increasing difficulty, the structure of the bottleneck, the supply and demand imbalance, the industry overall development rate of slow down. According to the industry development trend, the "cigarette market oriented reform" measures, to "demand forecast" as the end result, combined with the characteristics of cigarette marketing, the paper forecasted the demand for the next several years of the cloud, to provide a reference for the development of Yunnan tobacco company in the development of strategic planning, target decision making and the development of cigarette task allocation.First, this thesis expounded the concept, category and function of the forecasting theory knowledge, described the development and used of domestic and international forecasting methods, and further compared the advantages and disadvantages of different forecasting models, and analyzed the combination forecasting model Secondly, analyzed the performance of the cigarette market in the domestic cigarette market, the degree of competition, the brand status, the one or two kinds of cigarette sales trend, etc. Analyzed the development of the industry and the market oriented reform, concluded that the cigarette demand forecast is imperative, and will become the core and focus of future cigarette marketing work. Next, through the analysis of the importance, the economy and other factors to find out the impact of the eight elements of cloud cigarette prediction. Combining objective conditions, collected by Xinhua news agency, Yunnan branch, State Tobacco Monopoly Bureau, engineering system, State Statistical Bureau of the people’s Republic of China and other reference data. SPSS software was used to carry out stepwise regression analysis. The three factors that were not to be tested were the most closely related to cigarette sales. Because the model was not sensitive to the objective effects of the factors, such as the policy of the cigarette, and the major technical innovation, these factors are very important to the cigarette sales. For this reason, this thesis used the "two exponential smoothing method" to increase the objective conditions, such as periodic fluctuation and environmental change, and then forecast the cigarette sales again, and compared the results of two different forecasting methods, and found that the "multiple regression model" was used to the historical data, but the lack of objective factors, such as environment, policy and so on.The "two exponential smoothing method" can fully consider the objective effect, but the use of data information was not as "multiple regression model".The use of the two methods, can be used to compensate for each other, to avoid defects, so this thesis choosed "combination forecasting model" for cigarette sales forecast, analysis found that the forecast results of this method was close to the trend of the future sales.Forecast accuracy was also high, the future of cigarette sales, product planning, etc. Finally, the thesis analyzed the opportunities and challenges of the future national policy direction:the industry development environment, the Yunnan cigarette development prospect, and the potential challenges. According to the forecast results, the paper put forward suggestions and opinions on the cigarette marketing operation, the promotion of the market operation, the improvement of the cigarette value, the optimization of resources and the innovation of science and technology in four aspects.