Analysis And Empirical Study Of The Spatial Dependency Of Air Pollution

As one of key research fields of econometrics,spatial modeling techniques are the primary tool used by many scholars to study a variety of hot issues.Currently,in the context of environmental economics and financial economics,the frequent bilateral economic and trade exchanges strengthen the closeness of the cross-region linkages,irrespective of whether the geographies of regions are far apart or near.The flow of these economic representations implies the flow of environmental resources or financial risk assets.It has attracted the attention of this research on the spatial economic dependence effects of environmental measures and financial assets.This dissertation takes the dependency of carbon dioxide?CO₂?emissions,housing price indies and stock indies in China as the research object,focusing on the spatial economic structure of spatial econometric model.Through three empirical analyses,the spatial correlation theory based on economic distance is used to capture the characteristics of spillover effects among economies,and to explore the spatial dependence representations among regional housing markets with similar economic or air pollution conditions.First,this dissertation calculates CO₂ emissions intensity by selecting panel data from30 provinces,municipalities and autonomous regions in China over the period of1995-2013.This research integrates spatial interdependence in the EKC to consider the relationship between GDP per capita andCO₂ emission intensity,and examine to what extent CO₂ spillover depends on provinces'bilateral geographical and economic linkages.The results show that there is an inverted N-curve relationship betweenCO₂ emission intensity and GDP per capita,which shows a strong correlation against the effectiveness of the EKC hypothesis.Meanwhile,significant results show that economic linkage significantly outperforms all the other linkages in capturing the spatial dependencies between provinces.Second,this dissertation introduces a threshold effect regression model to cluster the30 provinces of China into different regions.Different subgroups are built to estimate and compare the degree of spatial interactions among regions under five definitions of spatial dependency,to study heterogeneous characteristics of the spatial dependency in the CO2emissions EKC.The results show that either before or after grouping,there is an inverted N-curve EKC relationship,and the spatial dependence of CO₂ emissions in the high-energy structure group is far greater than that in the low-energy structure group.Our results suggest that even though considering the spatial aggregation effect,the economic relations behind CO₂ emissions can also capture the spatial features more accurately.Third,this dissertation supplements the data of the past four years?2014-207?after the environmental policy of the Air Pollution Prevention Action Plan issued by the State Council in 2013.Using the panel SAR models with different time windows,this dissertation estimates recursively whether the spatial dependence characteristics of CO₂emissions are time-varying or not.The results show that the spatial dependence of CO₂emission EKC decreases gradually with time,irrespective of the specification of matrix form.The spatial correlation measure based on economic distance is most strongly attenuated and is stronger than other alternative definitions of the spatial correlation.Fourth,this dissertation uses spatial econometric models to study spatial dependency in price indices across 10 China housing markets.Selecting the new housing price indices from 2005 through 2016,this dissertations finds that geographical linkage is not the only factor that affects the spatial correlation across regional residential markets.After controlling the geographical closeness,the proximity of economic development or pollution level among regions has a significant impact on the spatial correlation across real estate markets.The studies find that spatial dependence can be stronger,especially for regions with far-reaching geographic locations but similar economic or air pollution conditions.Finally,using the dynamic spatial econometric model,this study examines the spatial spillover effect of the return series on 31 major stock index markets from period of 28st December 2012 to 28th June 2014.Parameter estimation is performed by using Clayton Copula model to construct the economic distance,which can measure the asymmetric tail dependence between two sets of financial return data.It can be reflect the consequences of extreme events.The results show that the Clayton-Copula connection function is used to estimate Kendall rank correlation coefficient to define the spatial weights matrix,which can reflect the extreme events contagion effect.the spatial correlation will be greater in the environmental policy strengthening stage.At the same time,a larger spatial correlation between stock indices is confined to regions with low CO₂emissions.Overall,it can be seen from the empirical analysis that the method proposed in this dissertation has the following innovations:?1?This dissertation theoretically points out that the bilateral economic relations are an important influence factor for the spatial dependence of CO₂ emissions.So far,the research about the spatial dependence with economic distance is still rare.?2?In empirical research,this dissertation extends the spatial contagion effect in the geographic and economic dimensions.this dissertation verifies that the spatial neighborhoods described by bilateral economic relations are far more superior in terms of capturing CO₂ spatial dependence.?3?Meanwhile,the study also considers the effect of the regional clustering about geographic and economic opportunities.This dissertation enriches the empirical results of the spatial dependence of CO2 emissions.?4?The study extends the research on the spatial correlation of real estate markets that can be stronger among different spatial regions with similar economic or pollution levels.