Research On Prediction Model Of Cigarette Sales Of Heilongjiang Tobacco Industrial Co.,LTD

In recent years,China’s tobacco industry is comprehensively promoting the work of "customer orders as the base of the organization of supply".This is also to meet today’s idea of "to sell production" in the society.Therefore,how to predict the sales accurately and provide real and effective reference and decision basis for the tobacco market operators has become a very important research topic.This paper introduced the grey system and the knowledge of the basic theory of Markov chain,analyzed the Grey Prediction Model and Markov Prediction Model of characteristic and scope of application: Grey Prediction Model is built using gray system theory.Gray Forecast Model required less information,ease of calculation,high prediction accuracy.But mainly for small data modeling has better accuracy.And Gray Forecast Model for predicting short time,less data,with a clear upward trend in the data series forecasting effect,when the random fluctuations in the data sequence are relatively la rge,the prediction accuracy will be significantly reduced.The study of Markov chain theory is a random dynamic system.More suitable for those who predict a relatively large stochastic fluctuations in stochastic processes,and Markov prediction model is more suitable for a large sample of data to predict.But the drawback is that the model required not only an object of Markov prediction,but also asked to predict the data sequence to be subject to certain typical distribution.However.For systems with little information,it is difficult to determine what data to predict the sequence is subject to distribution.Therefore,in this paper,the combination of the Grey System Prediction and Markov Prediction is reasonable,and the advantages of both can complement each other.The Gray Prediction Model predicted the general trend of the data sequence,and on the basis of data on trends in processing,the paper applied to the Markov Prediction Model,then the Gray Markov Prediction Model was formed.This paper studies the current situation of Hei Longjiang Tobacco Industry limited liability company sales forecasting mechanism.The analysis and summary is based on 36 monthly sales data in the total of 2012-2015 four years.Gray model,Markov model and the combination of two models,through the gray system theory and its application modeling software,the forecasting and analysis of the sample data and preparation of the corresponding procedures for Longjiang cigarette sales forecasting calculation.The analysis of the forecast results shows that this sales forecasting emphasis on analysis of sales data.In Heilongjiang Province,the annual sales volume and cigarette structure sales in the calculation of good results,the prediction of Heilongjiang Tobacco Industry limited liability company sales have played a positive role in promoting.