China Auto Possession Prediction Research

Since the car into our life, it developed very quickly and gradually becomes one of important consumer goods of our life. The development of the car brings a lot of convenience to our life, promoting the improvement of living standards, at the same time, the development of the car industry not only drive the upstream industry, but also the downstream, in a word, the progression and development of the auto industry plays an inestimable role in our national economy. The car industry brings us benefits, also generates a series of adverse effects, the rapid development of the automotive industry which makes energy and environmental problems become increasingly prominent, the exhaustion of the energy, the environment destruction and traffic problems and so on are all are we advocate for the sustainable development of society which can not be neglected. Therefore, the social from all walks of life produced great attention for the car consumption and its cause traffic problems. How to understand the prospect of the development of China's automotive industry, how to understand the car produces the favorable and unfavorable influence for economic development and society is one of the hot issues in current research. The research of Auto possession is a basic work in city planning and social development, how to accurately predict auto possession is significant for our urban development, transportation planning and the sustainable development. in view of this, this article adopts the improved traditional Multiple- linear regression, and put forward based on principal component analysis and a co integration element predictive regression model to research the factor and prediction of auto possession, which can produce a certain significance to our country, the healthy development of the auto industry and urban planning.Firstly, the paper summarizes and expounds research literature about the affecting factors and prediction of auto possession at home and abroad, the main methods on the affecting factors are the principal component analysis, factor analysis and other methods, the main methods on the forecast for the car are the neural network forecast method, the combination forecast method, the grey correlation method and so on. In a word, different people use different methods to analyze the affecting factors and the prediction of auto possession, which conclude different method has its advantages and disadvantages on the analysis of the auto possession, so it is very important to choose its own method on the basic of summarizing the predecessors' research.Secondly,the paper describes the function of the car industry in China's national economy and the present situation of the auto possession, so the research for auto possession provide certain theoretical and practical significance.The paper have chosen total eleven factor indexes which are per capita gross domestic product, industrial output, the national income, fixed assets investment, exports, society into total retail sales of consumer goods, population, urbanization rate, consumption level, energy consumption amount, highway mileage to forecast the auto possession, and set up the based on principal component analysis co integration regression forecast model. first to do the related analysis and principal component analysis on the indexes, the related analysis concludes the correlation among the dependent variable, at the same time, conclude there exist good correlation between the dependent variable and the independent variable, eliminate all of the correlation between the dependent variable, further uses the principal components analysis, with the main component replacement all the indexes have a comprehensive index, use the comprehensive index after the normalization process as a factor of forecasting the auto possession, then set up the single- linear regression model between the comprehensive index and the auto possession, before setting up the model, it needs to do unit root test and co integration test between the auto possession and the comprehensive index in order to eliminate the established model returns false. Using the co integration established model to forecast the auto possession, and the paper uses the fitted curve method to get the integrated index forecast equations to forecast the comprehensive index.On the basic outcome of the affecting factor and prediction about the auto possession, puts forward some policy Suggestions aimed at some energy, transportation and environmental issues which caused by the increase of auto possession and the development of the car industry.This article features place is put forward based on principal component co-integration regression model forecast method, the areas which need to be improved is the main process of prediction is to choose some quantifiable and available data in the prediction research, overlooked some policy and other factors, not comprehensive, should be further improved.