Dissecting the nonlinear economic implications of urban extreme thermo-environment using a Monte Carlo simulation-based ensemble learning model

• Published in: Habitat international Vol. 156; p. 103274
• Main Authors: Xiaochang, Liu, Renlu, Qiao, Xiuning, Zhang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.02.2025
• Subjects: Housing prices; Monte Carlo simulation-based ensemble learning model; Nonlinear interplay; Thermal implicit economic values; Urban extreme thermo-environment; Thermal implicit economic values; Urban extreme thermo-environment; Nonlinear interplay; Monte Carlo simulation-based ensemble learning model; Housing prices
• ISSN: 0197-3975
• DOI: 10.1016/j.habitatint.2024.103274

Numerous scholarly inquiries have indicated tangible urban amenities including structural and locational variables potentially contribute to housing prices, but intangible urban detriments characterised by environmental indices and their implicit economic implications remain largely unexplored. Escalating extreme heat events in cities, it is essential to delve into the observation of dissecting the nonlinear interplay between urban extreme thermo-environment (UETE) and housing prices. Consequently, we developed a theoretical framework to dissect the impacts of UETE on housing prices. Taking Shanghai as a testbed, UETE via percentile-based methods was estimated using remote sensing data from an intra-urban perspective. Moreover, a Monte Carlo Simulation-based interpretable ensemble learning from a hedonic perspective was proposed to examine its ramifications on housing prices. The results revealed the heterogeneous economic implications of hedonic determinants. The structural and locational variables, i.e., a south-facing orientation, middle floor levels, equipped elevators, proximity to the city centre, public transportation accessibility, prosperous economic activities and high-quality school districts exert positive impacts on housing prices whereas the environmental factors such as UETE exhibited negative influences on property values. For an increment of 1 °C in UETE from −3 °C to +0.0 °C, the property values will experience a decrease of 1250.15 RMB/m2. Once the mean UETE reaches the [+0.0 °C, +1.5 °C] threshold range, there is a significant diminishing marginal utility because of the trade-offs between environmental performance and property values. Intriguingly, as the mean UETE surpasses the +1.5 °C, the property values encounter a second reduction. Furthermore, the differences between housing prices inside UETE of varied typologies and those outside UETE were explored. The heterogeneous implicit economic implications of UETE underscore the necessity of making policies adapted to the unique idiosyncrasies of individuals, to alleviate the severe challenges of climate change towards the housing market.