Based On Panel Data Econometric Study On Economic Growth Effect Of Industry In China

Over the years,economists have done a lot of research on economic growth disparity.In the past studies,most economic units were treated and analyzed as individuals,ignoring the relevance of economic units.With the rapid development of economy,science and technology in recent years,especially the high-speed rail,the emergence of the emerging things such as 4G network has brought the expansion of regional openness and the rapid flow of knowledge,technology and elements,all these changes have made distance is no longer constraint on development,and economies also show more and more obvious accumulation effect.In the composition of GDP,the industrial economy is crucial.Exploring the influence of industrial economy on GDP is of great significance for predicting the level of China's economic development.In order to study the growth law of China's economy,this paper starts with the impact of industrial economy on GDP and carries out modeling and analysis of industrial economy and GDP data of all regions in China.First,from the perspective of spatial development,using the spatial econometric analysis method to detect spatial correlation between the economic variables,including 0-1 weights,co-movement spatial weights of space effect test.Under the condition of existing spatial correlation,a spatial cross-section model is established to explore the characteristics and rules of economic development in different years.Panel data is used to establish a spatial panel model,in order to explore the influence of space effect for China's overall economic development.Secondly,from the perspective of time,want to study the overall development trend of China's industrial economy,predict the distribution of economic development level and to explore the difference of regional economic development in China,it is necessary to use the method of interval symbol data analysis.Due to this article need to deal with interval data with known scatter,this article focuses on the internal known scatter interval symbol data analysis method.According to the characters of data found inside a scatter distribution is asymmetric,the analysis of the existing methods have limitations.Therefore,this paper put forward a way to use the median of interval data,upper value and lower value,as well as upper quartile and lower quartile of interval data.By using monte carlo simulation,this paper proves that the new method of least squares estimate shows superiority in some indicators significantly with the radius of the interval isindependent of the midpoint and does not change with time.When the interval radius independent of the midpoint and changes with time,the new model based on the least square estimation cannot show significant superiority.In order to improve the accuracy of model prediction,this paper adopts bayesian method to estimate parameters for the new model,and the results show that the new model under bayesian estimation is significantly superior to the existing model,and the empirical analysis results also prove that the new model has practical significance.Through the above two analysis methods to analyze the growth effect of China's industrial economy,a more comprehensive reflection of the law of China's industrial development on economic forecasting has practical significance.