南京市住宅价格影响因素分析
秦佳睿(1997— ),女,河南洛阳人,硕士生,主要从事城市与区域经济研究。E-mail:qinjiarui1997@foxmail.com |
收稿日期: 2020-06-05
要求修回日期: 2020-08-15
网络出版日期: 2021-07-25
基金资助
国家自然科学基金项目(41631175)
国家重点研发计划项目(2017YFB0503500)
版权
Analysis of the Influencing Factors of Housing Price in Nanjing
Received date: 2020-06-05
Request revised date: 2020-08-15
Online published: 2021-07-25
Supported by
National Natural Science Foundation of China(41631175)
National Key Research and Development Program of China(2017YFB0503500)
Copyright
住宅价格影响因素的研究对于购房者购买住房及政府制定相关政策具有重要影响。然而,目前对于住宅价格问题的研究大多从宏观角度进行,而从微观角度入手的相对较少。此外,对住宅价格的研究通常需要建立多元回归方程,但这样会造成多重共线性问题,导致伪回归,不能准确地分析各因素对住宅价格的影响。针对该问题,本文从微观角度出发,运用定量与定性相结合的方法,建立特征价格模型,并且选择逐步回归法对其进行修正,将其中显著性不强、经济意义不明显的变量逐步剔除,从而更准确地探究影响住宅价格的主要因素。本文以南京市住宅价格为实验数据,并初步选择12个影响因素对该方法进行验证。结果表明,该方法能够有效地剔除二级商业中心、建筑类型、生活配套设施3个显著性不强的影响因素,保留重点学校等9个影响较大的因素。本文方法更为精准地分析了住宅价格的影响因素,为购房者购买住房及政府制定相关政策提供了一定的理论基础。
秦佳睿 , 盛业华 , 王燕锋 , 何育枫 . 南京市住宅价格影响因素分析[J]. 地球信息科学学报, 2021 , 23(5) : 882 -890 . DOI: 10.12082/dqxxkx.2021.200285
With the urbanization and the rapid growth of economy, the housing price of cities continues to rise and the spatial differentiation of housing price increases, which has attracted a widespread attention from the public and the government. Exploring the influencing factors of housing price is of great significance for understanding house-buying of the public and provides useful support for the government to formulate relevant policies. Until now, many researches have studied the influencing factors of housing price from a macro perspective and only a few studies are conducted from the micro perspective, which leads to inadequate understanding of the influencing factors of housing price. In addition, the method of establishing multiple regression equations is usually adopted to study the relationship between housing price and multiple influencing factors. However, the selection of influencing factors are usually subjective, which leads to an artificial regression that cannot characterize the influence of various factors accurately. To solve these problems, a hedonic price model combining both quantitative and qualitative methods is proposed from the micro perspective of structure characteristics, location characteristics, and neighborhood characteristics in this study. Moreover, a stepwise regression method is applied to modify this model by eliminating non-significant factors. To validate the proposed method, a case study in Nanjing city, Jiangsu is conducted, and 12 primary influencing factors are selected. The results show that the proposed method eliminates three insignificant influencing factors including secondary commercial centers, building types, and living facilities, and keeps the rest nine factors. Compared with traditional methods, this paper analyzes the influencing factors of housing price more quantitatively and provides a theoretical basis for house-buying of the public and policy making of the government. What's more, it is conducive to the real estate management department to have an in-depth understanding of the distribution of housing price, so as to strengthen the management of the real estate market.
表1 住宅特征变量的描述与量化Tab. 1 Description and quantification of residential characteristic variables |
变量类型 | 特征变量 | 变量描述 | 量化表达 |
---|---|---|---|
区位特征 | 地铁站距离(Subway) | 与最近地铁站距离 | 1 km内赋为5, 1~2 km赋为4, 2~3 km赋为3, 3~5 km赋为2, 5 km以上赋为1 |
公交站点距离(Bus Station) | 与最近公交站点距离 | 50 m内赋为5, 50~100 m赋为4, 100~500 m赋为3, 500~1000 m赋为2, 1000 m以上赋为1 | |
城市中心 (Urban Center) | 与南京市新街口的距离 | 5 km内赋为1, 5 km以上赋为0 | |
二级商业中心(Second Center) | 各行政区内的二级商业中心,如河西、城南、城南路、东山等 | 1 km内赋为4, 1~3 km内赋为3, 3 ~5 km内赋为2, 5 km以上赋为1 | |
建筑特征 | 绿化率(Green) | 小区绿化面积/小区规划面积 | 小区绿化率 |
物业费用(Cost) | 住宅小区的物业管理费用(元/月/m2) | 具体数值 | |
住宅房龄(Age) | 住宅建成的时间 | 当前时间与建成时间差值 | |
建筑类型(Style) | 低层、多层、小高层、高层、超高层 | 低层赋为1,多层赋为2,小高层赋为3,高层赋为4,超高层赋为5 | |
邻里特征 | 生活配套设施(Market) | 与最近大型超市、商场的距离 | 200 m内赋为5, 200~500 m赋为4, 500 m~1000 m赋为3, 1~3 km赋为2, 3 km以上赋为1 |
医疗设施(Hospital) | 与最近的三甲医院的距离 | 500 m内赋为5, 500~1 km赋为4, 1~3 km赋为3, 3~5 km赋为2, 5 km以上赋为1 | |
教育配备(Key Primary) | 与重点小学的距离 | 根据政府实际文件,位于重点学区内的赋为1,反之赋为0 | |
休闲设施(Park) | 与最近的城市公园的距离 | 1 km内赋为5, 1~2 km赋为4, 2~3 km赋为3, 3~5 km赋为2, 5 km以上赋为1 |
注:城市中心5 km内考虑作为新街口重点辐射区。 |
表2 特征价格模型描述Tab. 2 Hedonic price model description |
模型 | 包含的特征变量 |
---|---|
1 | Key Primary |
2 | Key Primary、Cost |
3 | Key Primary、Cost、Urban Center |
4 | Key Primary、Cost、Urban Center 、Subway |
5 | Key Primary、Cost、Urban Center 、Subway、Bus Station |
6 | Key Primary、Cost、Urban Center 、Subway、Bus Station、Green |
7 | Key Primary、Cost、Urban Center 、Subway、Bus Station、Green、Park |
8 | Key Primary、Cost、Urban Center 、Subway、Bus Station、Green、Park、Hospital |
9 | Key Primary、Cost、Urban Center 、Subway、Bus Station、Green、Park、Hospital、Age |
表3 特征价格模型的显著性检验Tab. 3 Significance test of the hedonic price model |
模型 | R | R2 | 调整后的R2 | 标准估计的误差 | 更改统计 | ||||
---|---|---|---|---|---|---|---|---|---|
R2变化量 | F变化量 | 自由度1 | 自由度2 | Sig.F 变化量 | |||||
1 | 0.612 | 0.374 | 0.374 | 0.3768 | 0.374 | 1854.534 | 1 | 3100 | 0.000 |
2 | 0.641 | 0.411 | 0.411 | 0.3655 | 0.037 | 195.276 | 1 | 3099 | 0.000 |
3 | 0.669 | 0.447 | 0.447 | 0.3542 | 0.036 | 201.154 | 1 | 3098 | 0.000 |
4 | 0.685 | 0.470 | 0.469 | 0.3470 | 0.023 | 132.067 | 1 | 3097 | 0.000 |
5 | 0.697 | 0.486 | 0.485 | 0.3418 | 0.016 | 94.619 | 1 | 3096 | 0.000 |
6 | 0.704 | 0.496 | 0.495 | 0.3385 | 0.010 | 63.119 | 1 | 3095 | 0.000 |
7 | 0.709 | 0.503 | 0.502 | 0.3362 | 0.007 | 42.205 | 1 | 3094 | 0.000 |
8 | 0.711 | 0.505 | 0.504 | 0.3355 | 0.002 | 14.328 | 1 | 3093 | 0.000 |
9 | 0.712 | 0.507 | 0.506 | 0.3352 | 0.001 | 7.090 | 1 | 3092 | 0.000 |
表4 回归系数t检验、F检验和多重共线性检验Tab. 4 Regression coefficient t test, F test and multi collinearity test |
模型 | 自变量 | t | Sig.t | F | Sig.F | 共线性统计 | |
---|---|---|---|---|---|---|---|
Tolerance | VIF | ||||||
9 | Key Primary | 24.126 | 0.000 | 1854.534 | 0.000 | 0.534 | 1.872 |
ln(Cost) | 13.922 | 0.000 | 2.98 | 0.000 | 0.643 | 1.556 | |
Urban Center | 10.249 | 0.000 | 1059.049 | 0.000 | 0.552 | 1.812 | |
Subway | 8.857 | 0.000 | 126.168 | 0.000 | 0.830 | 1.205 | |
Bus Station | 8.899 | 0.000 | 204.262 | 0.000 | 0.762 | 1.313 | |
ln(Green) | 8.749 | 0.000 | 1.308 | 0.018 | 0.871 | 1.148 | |
Park | 4.873 | 0.000 | 91.752 | 0.000 | 0.759 | 1.318 | |
Hospital | 3.725 | 0.000 | 107.512 | 0.000 | 0.657 | 1.523 | |
ln(Age) | 2.754 | 0.001 | 7.438 | 0.000 | 0.515 | 1.942 |
表5 模型9回归系数表Tab. 5 Table of regression coefficients |
自变量 | 非标准化系数 | |
---|---|---|
B | 标准误差 | |
(常量) | 9.482 | 0.055 |
Key Primary | 0.405 | 0.017 |
ln(Cost) | 0.147 | 0.011 |
Urban Center | 0.173 | 0.017 |
Subway | 0.057 | 0.007 |
Bus Station | 0.062 | 0.007 |
ln(Green) | 0.163 | 0.019 |
Park | 0.040 | 0.008 |
Hospital | 0.030 | 0.008 |
ln(Age) | 0.030 | 0.011 |
表6 模型9各特征变量的价格弹性系数Tab. 6 The price elasticity coefficient of each characteristic variable |
影响因素 | 非标准化系数 | 弹性系数/% | 半弹性系数/% |
---|---|---|---|
ln(Green) | 0.163 | 0.163 | |
ln(Cost) | 0.147 | 0.147 | |
Urban Center | 0.173 | 18.887 | |
Key Primary | 0.405 | 49.930 | |
Bus Station | 0.062 | 6.396 | |
Subway | 0.057 | 5.865 | |
Park | 0.040 | 4.081 | |
Hospital | 0.030 | 3.045 | |
ln(Age) | 0.030 | 3.045 |
[1] |
王芳, 高晓路, 颜秉秋. 基于住宅价格的北京城市空间结构研究[J]. 地理科学进展, 2014,33(10):1322-1331.
[
|
[2] |
梅志雄, 徐颂军, 欧阳军, 等. 广州地铁三号线对周边住宅价格的时空影响效应[J]. 地理科学, 2011,31(7):836-842.
[
|
[3] |
宋伟轩, 毛宁, 陈培阳, 等. 基于住宅价格视角的居住分异耦合机制与时空特征——以南京为例[J]. 地理学报, 2017,72(4):589-602.
[
|
[4] |
温海珍, 李旭宁, 张凌. 城市景观对住宅价格的影响——以杭州市为例[J]. 地理研究, 2012,31(10):1806-1814.
[
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
许基伟, 马欣. 基于Hedonic模型的小区房价影响因素研究——以南京市鼓楼区为例[J]. 工程经济, 2020,30(1):61-64.
[
|
[12] |
刘俊豪. 住宅特征对城市二手房挂牌价的影响研究——以徐州市为例[D]. 徐州:中国矿业大学, 2016.
[
|
[13] |
南京市规划和自然资源局. . 《南京市城市总体规划(2018—2035)》草案[EB]. http://ghj.nanjing.gov.cn/njsgtzyj/201812/t20181225_1347362.html.
[ Nanjing Bureau of Planning and Natural Resources. Draft of "Nanjing City Master Plan (2018-2035)"[EB]. http://ghj.nanjing.gov.cn/njsgtzyj/201812/t20181225_1347362.html. ]
|
[14] |
尹上岗, 李在军, 宋伟轩, 等. 基于地理探测器的南京市住宅租金空间分异格局及驱动因素研究[J]. 地球信息科学学报, 2018,20(8):1139-1149.
[
|
[15] |
钟芬. 棕地对周边住宅价格的影响研究[D]. 合肥:合肥工业大学, 2017.
[
|
[16] |
刘倩倩. 基于3C定价理论的上海商品住宅定价策略的实证研究[D]. 上海:华东师范大学, 2012.
[
|
[17] |
刘立祥. 线性回归模型中自变量的选择与逐步回归方法[J]. 统计与决策, 2015(21):80-82.
[
|
[18] |
游士兵, 严研. 逐步回归分析法及其应用[J]. 统计与决策, 2017,14:31-35.
[
|
[19] |
彭保发, 石忆邵, 单玥, 等. 上海市三甲医院对周边地区住房价格的空间影响效应分析[J]. 地理科学, 2015,35(7):860-866.
[
|
[20] |
石忆邵, 王伊婷. 上海市学区房价格的影响机制[J]. 中国土地科学, 2014,28(12):47-55.
[
|
[21] |
|
[22] |
董藩, 董文婷. 学区房价格及其形成机制研究[J]. 社会科学战线, 2017(1):43-51.
[
|
/
〈 | 〉 |