Application Of Grey Forecasting Model In The Consumption Structure Of Chongqing City

As a kind of social practice activity, it is produced and developed in order to adapt to the development of social economy. Scientific prediction method has become an important application of various departments, and plays an important role in the national economy. Grey system theory is a method, studying on the non-determinism system that has characteristic of some information having known while others remain unknown, it is created by China's famous scholar Professor Deng Julong. Grey forecasting model is its important content, and grey forecasting is a predication method which is widely used. In modeling conditions and the choice of sample volume, it has many advantages, such as strong flexibility, convenient operation, short-term prediction of high precision and testable, but it exists many limitations likes the other prediction methods, in the long term, it's prediction accuracy is low, so on the basis of the basic method, in order to get good prediction effect, we will apply the improved model in the empirical analysis.In this paper, we first introduce the related concepts of the grey system theory and the basic knowledge of the grey prediction model, then the cross section data of the per capita consumption expenditure in 2000 to 2014 were analyzed by cluster analysis using software SPSS. The structure of consumption is carried on grey correlation degree analysis after the data are divided into four stages. In this paper, the consumer spending, tobacco and food, housing, clothing, education, culture and entertainment, transportation and communications and other factors in various stages of grey relational analysis by using R software, at the same time,the consumption structure of urban residents in Chongqing and the internal relation are analyzed in detail. Then forecasting analysis by using the grey forecasting model GM(1,1) model, the original sequence based on logarithmic transformation improved GM(1,1) model, the grey metabolic GM(1,1) model and a combination of the two models, from the practical proof that the prediction accuracy of the improved model is higher than traditional model, and obtained that the prediction accuracy of the combination forecasting model is the best, and to predict the consumer spending in the 2016 and 2020 by using the four methods. Finally, we put forward some problems about the consumption structure of urban residents in Chongqing, and then give the corresponding policy recommendations to improve the consumption structure.