Exploring the value of green: The impact factors on China's second-hand green housing prices based on geographically weighted Lasso regressions

Abstract

Green housing development has progressed over the past two decades; however, the pricing advantages and influencing factors remain inadequately defined. Particularly, the effects of green housing ratings on prices and purchasing decisions have not been thoroughly researched. A dataset comprising 14,335 items (4101 groups) was compiled of second-hand green housing transactions from spatial and temporal dimensions across various Chinese cities. To manage data clustering and enhance model robustness, the Bootstrap algorithm sampling and geographically weighted Lasso regression were utilized. The findings reveal several insights: (1) The spatial dimension notably impacts second-hand green housing prices, with regional differences evident in the effects of identical variables. This suggests that policy should be locally adapted, requiring nuanced and differentiated regulatory strategies. (2) Macroeconomic indicators, such as Gross Domestic Product, Per Capita Disposable Income, and residential commercial property sales, positively influence housing prices. Monitoring these economic indicators for timely policy adjustments is advised. (3) At the microeconomic level, the architectural features of second-hand green housing negatively affect prices in the northeast and southwest regions. Conversely, neighborhood characteristics negatively impact prices in the southeast coastal region but positively influence them in central and northeastern regions. These results suggest that regular assessments of neighborhood characteristics and stringent regulation of architectural features by the government are necessary to maintain housing stock quality. This research offers enhanced insights into price formation in the second-hand green housing market and presents vital evidence for precise policy formulation and sustainable real estate development.