Research On Influence Mechanism Of Housing Price,Economic Growth And Urbanization Based On Network Structure

From 2018 to July 2019,the political bureau of the central committee of the communist party of China(CPC)held six important meetings to address the problem of rising real estate prices in China.Accordingly,the academia has made extensive exploration and research on the relationship between real estate market price,economic growth and urbanization.Therefore,on the basis of existing research,further digging the impact mechanism between the real estate market price and economic growth and the urbanization rate not only have important academic value,but the research conclusions also have certain policy value.The main work of this paper is as follows.First,through literature review,the main variables affecting real estate prices were extracted based on existing studies,including GDP growth rate,urbanization rate,wage level,housing supply level and financial affordability.Second,Granger Causality Test is used to test the real estate market price,GDP growth rate and population urbanization rate in 253 Chinese cities.The results show that there is a one-way causal relationship between economic growth and housing price(housing price is the Granger cause of economic growth).There is a one-way causal relationship between housing price and urbanization rate(urbanization rate is the Granger cause of housing price).Population urbanization is the Granger cause of GDP growth.Third,based on the Granger causality test results,combined with economic theory,three Bayesian network structures with high probability are constructed,and the optimal Bayesian network structure is selected using the BIC score.The research results show that:(1)prefecture-level cities and provincial capital cities have the same optimal Bayesian network structure.The structure considers that financial affordability,urbanization rate and housing price are the parent nodes of economic growth,while doctors and wages are the parent nodes of urbanization rate.Housing supply and urbanization rate are the parent nodes of housing price,and construction land supply is the parent node of the housing supply.(2)The municipality and the planned city have the same optimal Bayesian network structure.The structure considers that financial affordability,urbanization rate and housing price are the parent nodes of economic growth,while the number of doctors and housing supply are the parent nodes of urbanization rate.The wage,construction land supply and urbanization rate are the parent nodes of housing prices.Fourth,according to the characteristics of different types of cities with different Bayesian network structures,Bayesian network learning and reasoning were carried out respectively to study the interaction mechanism of real estate market price,GDP growth rate and population urbanization rate.The research results for prefecture-level cities and provincial capital cities show that:(1)financial affordability has a negative impact on economic growth.(2)housing supply has a negative impact on housing prices.(3)the increase of population urbanization rate will greatly reduce the housing supply.(4)wages have a large impact on the urbanization rate of population.Due to the difference in wages,the probability of the urbanization rate of population within the same range varies by 0.071.(5)housing supply has a strong influence on the growth rate of housing price,and the change range of housing price growth rate in the same range is 0.054.The research results of municipalities directly under the central government and cities separately listed in the state plan show that:(1)financial affordability has a positive impact on economic growth.(2)housing supply has a negative impact on urbanization.(3)housing supply has a small impact on urbanization.Due to the different housing supply,the probability of population urbanization rate in the same interval changes only 0.1.(4)wages have a large influence on the growth rate of housing prices.Due to the difference in wages,the probability of housing price growth rate in the same range changes by 0.236.The innovation of this paper lies in the following two points:The first is to establish a Bayesian network between real estate prices,GDP growth rate and population urbanization rate by taking 235 Chinese cities as research objects and combining the results of Granger causality test and economic theory.By means of BIC scoring index,the results of three Bayesian networks with the greatest possibility were tested for different types of cities to determine the optimal Bayesian network structure corresponding to different types of cities.The results show that prefecture-level cities and provincial capitals have one optimal Bayesian network structure,while municipalities directly under the central government and cities with separate planning have another optimal Bayesian network structure.The second is to study the interaction mechanism of real estate price,GDP growth rate and population urbanization rate based on the optimal Bayesian network structure of different types of cities,using parameter learning and reasoning.The results show that the influence mechanism of real estate price,GDP growth rate and population urbanization rate is different for different types of cities.Based on the above conclusions,it is proposed that the government should implement real estate development planning and urban planning with regional differentiation,promote the coordinated development of China’s real estate industry and urbanization,adopt a stable and sustainable urbanization strategy,and implement a long-term real estate mechanism.