时空异质性探测的上海市房价演进分析  

The evolution analysis of Shanghai housing price based on spatiotemporal heterogeneity detection

作　　者：张俊杰 杨毅 蒋廷臣 徐胜华[2] ZHANG Junjie;YANG Yi;JIANG Tingchen;XU Shenghua(Jiangsu Ocean University,Lianyungang,Jiangsu 222023,China;Chinese Academy of Surveying and Mapping,Beijing 100036,China)

机构地区：[1]江苏海洋大学,江苏连云港222023 [2]中国测绘科学研究院,北京100036

出　　处：《测绘科学》2020年第12期175-180,196,共7页Science of Surveying and Mapping

基　　金：国家自然科学基金项目(41801316);江苏省高校自然科学研究面上项目(17KJD420001);江苏高校优势学科建设工程资助项目。

摘　　要：针对传统地理加权回归模型对城市房价演进及其影响因素的研究无法克服时空异质性问题,该文通过时空异质性判别对数据的非平稳性进行时空探测,将时间维度信息引入回归模型,提出了城市房价演进和影响因素分析方法。以上海市2000—2018年房价数据为研究对象,利用时空地理加权回归模型进行分析,得到各自变量影响因素的回归系数,同时利用克里金插值方法对系数矩阵进行分析。实验结果表明,采用时空地理加权回归模型,能够有效地解决城市房价的时空异质性问题,更好地揭示出不同特征因素对上海市房价影响程度的分布格局在时空维度上的变化规律。In view of the fact that the traditional geographically weighted regression model can not overcome the problem of spatialtemporal heterogeneity in the study of urban housing price evolution and its influencing factors,this paper explored the non-stationarity of the data through discrimination of spatiotemporal heterogeneity,introduced the time dimension information into the regression model,and put forward the analysis method of urban housing price evolution and influencing factors.Based on the housing price data of Shanghai from 2000 to 2018,the regression coefficients of the influencing factors of each variable were obtained by using the geographically and temporally weighted regression model,and the coefficient matrix was analyzed by Kriging interpolation method.The experimental results showed that the geographically and temporally weighted regression model could effectively solve the spatiotemporal heterogeneity of urban housing prices,and better reveal the spatialtemporal variation of the distribution pattern of different characteristic factors affecting Shanghai housing price.

关 键 词：时空地理加权回归 上海房价 时空异质性 影响因素 

分 类 号：P208[天文地球—地图制图学与地理信息工程]