地理信息系统及空间模型在区域发展中的研究-以北京为例 GIS and Spatial Modeling in Regional Development Studies Case Study on the Greater Beijing Area.docVIP

GIS and Spatial Modeling in Regional Development Studies: Case Study on the Greater Beijing Area Danlin Yu Ph.D. Candidate Department of Geography University of Wisconsin – Milwaukee Milwaukee, Wisconsin, 53201 Tel: (414) 229-5818 Fax: (414) 229-3981 Email: danlinyu@uwm.edu 1 INTRODUCTION GIS application in regional studies has recently attracted scholars’ interests. However, most of the studies focused on GIS’s mapping and monitoring functionalities. As a spatial data management system, its spatial analytical aspects remained largely unexploited. This study intends to integrate spatial modeling into GIS to investigate the regional development mechanisms in the Greater Beijing Area, China. In the literature, investigation of development mechanisms is primarily conducted through regression analysis. Non-spatial (OLS regression) or global spatial (spatial lag or error regression) approaches were most commonly employed. The non-spatial approach is best suited for larger scale study, such as provincial level spatial units, since high-level of data aggregation intends to mask subtle spatial effects among spatial units. The global spatial approaches, on the other hand, are able to handle the spatial effects, especially the spatial dependence by explicitly including a spatial factor (either a spatial lag or a spatial error) in the model construction and calibration. However, another closely related spatial effect, the spatial heterogeneity, is not fully addressed in the global spatial approaches. In addition, due to its global nature, the spatial regression only gives a general picture of development mechanisms in practice. Although GIS is closely integrated in developing a global spatial model for spatial information retrieving and manipulation, the calibration of the model cannot be returned to GIS for visualization. To this end, this study intends to employ a recently developed GIS alternative, the geographically weighted regression (GWR) to investigate the spatial het