城市化对房价的影响:线性还是非线性?——基于四种面板数据回归模型的实证分析

The Effect of Urbanization on House Prices:Linear or Nonlinear?——Empirical Study Based on Four Panel Data Regression Models

文章从线性和非线性两个角度分析了中国城市化进程对房价的影响。通过对各省历年房价和城市化的核密度估计空间分布分析,发现城市化和房价之间存在明显的正相关性,并且各省份的城市化和房价水平存在"双峰"分布特征和空间相关性。这说明在分析城市化对房价的影响时应考虑可能的门限效应和空间溢出效应这两种非线性关系。据此,文章基于中国30个省份1998-2009年的面板数据,使用普通面板回归、空间面板回归、门限面板回归和平滑门限面板回归这四种模型进行分析发现,城市化水平对本地区和相邻地区的房价均具有显著的促进作用,且在经济增长水平较高、人力资本集聚的地区,城市化对房价的促进作用更加显著。

This paper analyzes the effect of urbanization on house prices from the linear and nonlinear angles.Based on kernel density estimation and spatial distribution analysis of urbanization and house prices,it finds that urbanization significantly and positively correlates with house prices and provincial urbanization and house prices are featured by distinct bimodal distribution and spatial autocorrelation,which illustrates the consideration of two possible nonlinear relationship,namely threshold effect and spatial spillover effect,when analyzing the impact of urbanization on house prices.Thus,based on the panel data of 30 provinces from 1998 to 2009 and four models,namely ordinary panel regression,spatial panel regression,panel threshold regression and panel smooth threshold regression,this paper finds that the urbanization level has significant positive impact on house prices in local and adjacent regions,and the impact will be more significant in regions with higher level of economic growth and the concentration of human capital.