Study On The County Relative Spacitial Location Index In Urban Agglomeration

In recent years, urban agglomeration has gradually become a powerful form ofregional economic development. At the same time, as the basic administrative regionaleconomy, county economy is payed more and more attention. However, academics tend tosdudy them respectively, while paying little attention to the relationship. In fact, urbanagglomeration is composed of cities and counties, and county plays the role of basic unitin an urban agglomeration. The relationship between urban agglomeration and county isjust as the whole and the part. So the development of them both coexists, and promoteseach other. The spatial location of a county in the urban agglomeration is unique, which isdetermined by the relationships among the agglomeration’s elements. This kind oflocation influences the level of the technical efficiency of county, being the research focusof this dissertation. Based on the stochastic production frontier model, this dissertationmeasures and computates relative spatial location variables of county, and get a locationindex finally.First of all, this dissertation makes a comprehensive analysis of the researchbackground and status of urban agglomeration, county economic and technical efficiency,pointing out the fault zone of the research. Regarding this as the cutting point, this partcombines the theories and actual situation, and constructed the theoretical framework ofthis dissertation.Secondly, stochastic production frontier analysis method is selected as the main tools,with the theoretical explanation and description of model construction. The research objietis a six-year panel data set of769counties and161cities in23Chinese urbanagglomerations. Major part is the construction of the stochastic production frontier modelwith spatial variables, which calculates technical efficiency of counties.Thirdly, based on the structure relationships among urban agglomeration, city,county, this part designs a total of18spatial variables. Then, these variables are validatedby their variance inflation factors, and13of them are selected to be combianed in the nextstep. Every varialbe need to be introducted one by one, and finally be combined in the optimal form.Finally, a county relative spatial location measure index is constructed, through thespatial variable optimal form. Results of three inspections show that this index has asatisfying ability to describe the real situation of county location. Considering the form ofthis index, this dissertation proposes target-oriented action suggestions to optimize thecounty relative spatial location.