中文摘要：

用传统统计学方法探讨城市地价空间分布与变化的前提条件是用于分析的地价数据在统计上必须独立且均匀分布，但是地价数据在空间上一般存在一定的空间依赖关系（空间自相关），同时这些空间关系中也隐藏着某些有用的信息，分析这些隐藏信息可以挖掘城市地价模型在微观层面上的适用性．论文以南京市主城区作为研究区域，采用Moran’S Ⅰ和Local Moran’s Ⅰ系数来表示城市地价的空间自相关性特征，建立了城市地价影响因子的经典线性回归模型和空间A回归模型，并比较了这两种模型的分析结果．研究结果表明：南京市住宅、工业、商业地价均具有空间正相关性与空间集聚特征，住宅地价与工业地价的空间集聚特征较商业地价更为明显，且呈上升趋势．空间自回归模型的拟合度和解释能力要优于经典线性同归模型，且其残差的空间自相关性消失，但是其自变量的显著水平比经典线性回归模型低．

英文摘要：

Land price drivers that best describe land price spatial distribution quantitatively are often selected through regression analysis. A problem using conventional statistical methods in spatial land price analysis is that these methods assume the data to be statistically independent while land price data have the tendency to be dependent spatially, known as spatial autocorrelation. By mining these concealed spatial autocorrelation information, we can explore the feasibility of applying urban land price model in a microscopic view. As a result, spatial statistical method was applied in this study to derive the spatial distribution of land price. In this paper, Nanjing city was selected as the study area. Moran＇s Ⅰ and local Moran＇s Ⅰ are used to describe spatial autocorrelation of land price. What＇s more, standard linear regression model and spatial autoregressive model of land price are constructed. Results show that residential land price, industrial land price, and commercial land price all have spatial autocorrelation characteristics. Compared with that of industrial land price and commercial land price, spatial aggregation characteristics of residential land price are higher, and the trend of spatial aggregation becomes more and more obvious. The spatial autoregressive model is statistically sound in the presence of spatially dependent data in contrast with the standard linear model, and it has a better goodness of fit. At the same time, spatial autocorrelation in the residuals of spatial autoregressive land price model has disappeared. By using spatial models a part of the variance is explained by neighboring values. The estimated regression coefficients of the variables become smaller and the significance of the parameters also decreases in the spatial autoregressive model.