| 基于街景图像的厦门本岛环境特征对住宅价格的影响研究 |
| |
| 引用本文： | 郭金函,马子迎,边经卫,江楚钶. 基于街景图像的厦门本岛环境特征对住宅价格的影响研究[J]. 地球信息科学学报, 2022, 24(11): 2128-2140. DOI: 10.12082/dqxxkx.2022.220538 |
| |
| 作者姓名： | 郭金函 马子迎 边经卫 江楚钶 |
| |
| 作者单位： | 华侨大学建筑学院，厦门 361021 |
| |
| 摘 要： | 住宅价格的空间分异是城市空间资源配置不均衡的外在表现，理解住宅价格的主导影响因素及其空间分异特征对于住区规划及房价调控政策的制定具有重要意义。既有研究较少考虑环境品质对住宅价格的影响和影响因素的作用尺度差异，针对以上问题，本文引入街景图像，在特征价格模型的基础上拓展环境特征，构建多尺度地理加权回归（MGWR）模型，研究环境特征对住宅价格的影响效用，并通过分析其他控制变量的系数空间格局，总结各变量的空间分异特征规律。主要结论为：(1)街景图像测度的环境特征更符合人们对居住环境的真实感知，研究结论可为居住环境品质提升提供更加精细化的设计策略；MGWR模型对变量的空间分异现象具有更接近于真实值的拟合效果，可描述不同变量的作用尺度差异，这有助于为特定地区制定针对性规划策略。(2)厦门本岛住宅价格呈现显著的聚类特征，并沿城市核心发展轴呈“带状”结构分布。(3) 3个环境特征变量对于住宅价格均为显著的正向影响，且作用接近全局尺度，街景绿视率的影响最强，其次是天空开敞度和相对步行指数。(4)总结各变量的系数空间分异规律，发现不同特征地区住宅价格的主导影响因素不同，核心地区主要受交通、教育因素的影响；...
|
| 关 键 词： | 住宅价格 街景图像 环境特征 多尺度地理加权回归 特征价格模型 影响因素 厦门 |
| 收稿时间： | 2022-07-23 |
Analysis on the Influence of Environmental Characteristics of Xiamen Island on Housing Price based on Street View Imagery |
| |
| GUO Jinhan,MA Ziying,BIAN Jingwei,JIANG Chuke. Analysis on the Influence of Environmental Characteristics of Xiamen Island on Housing Price based on Street View Imagery[J]. Geo-information Science, 2022, 24(11): 2128-2140. DOI: 10.12082/dqxxkx.2022.220538 |
| |
| Authors: | GUO Jinhan MA Ziying BIAN Jingwei JIANG Chuke |
| |
| Affiliation: | School of Architecture, Huaqiao University, Xiamen 361021, China |
| |
| Abstract: | The spatial differentiation of housing price is an external expression of the unbalanced allocation of urban spatial resources. It is of great significance for formulation of residential planning and housing price control policies to understand the leading influencing factors and their spatial differentiation characteristics. Existing studies seldom consider the impact of environmental quality on housing price and the different scale effects of influencing factors. In response to the above problems, the study introduces street view imagery, adds environmental characteristic variables based on the traditional Hedonic price model, and constructs Multiscale Geographically Weighted Regression (MGWR) model to explore the impact of environmental characteristics on housing price. The spatial pattern of coefficients of other control variables are analyzed and the spatial differentiation characteristics of each variable are summarized. The main conclusions are: (1) The environmental characteristics measured by street view imagery are more consistent with people's real perception of the living environment, and the research conclusions can provide more refined strategies for improving the quality of the living environment. The MGWR model has a fitting effect on the spatial differentiation of variables closer to the real value, and it can describe the difference of action scales of different variables, which is helpful to formulate targeted planning strategies for specific regions; (2) The housing prices on Xiamen Island show significant clustering characteristics, and are distributed in a "band" structure along the city’s core development axis; (3) The three environmental characteristics variables have significant positive effects on housing prices, and the effects are close to the global scale. The impact of street view green viewing rate is the strongest, followed by sky openness and relative walkability index; (4) By summarizing the spatial differentiation rules of the coefficients of each variable, it is found that the dominant influencing factors of housing price in different characteristic areas are different. The core areas are mainly affected by traffic and education factors; the old urban areas are mainly affected by environmental quality and building quality factors; the new urban areas are mainly affected by location and living facilities. |
| |
| Keywords: | housing price street view imagery environmental features multiscale geographically weighted regression model hedonic price model influencing factor Xiamen |
|
| 点击此处可从《地球信息科学学报》浏览原始摘要信息 |
|
点击此处可从《地球信息科学学报》下载全文 |
