Identifying multi-scale spatial heterogeneity in the urban housing market: A fusion approach combining random effects eigenvector spatially filtering-based spatially varying coefficient (RE-ESF-SVC) model with fused lasso
Topics: Spatial Analysis & Modeling
, Urban Geography
,
Keywords: spatial heterogeneity, spatial scale, real estate market, eigenvector spatial filtering, spatially varying coefficient model, fused lasso
Session Type: Virtual Paper Abstract
Day: Saturday
Session Start / End Time: 2/26/2022 09:40 AM (Eastern Time (US & Canada)) - 2/26/2022 11:00 AM (Eastern Time (US & Canada))
Room: Virtual 34
Authors:
Zhan Peng, Tohoku University
Ryo Inoue, Tohoku University
,
,
,
,
,
,
,
,
Abstract
Due to regional differences, structure, accessibility, and neighborhood factors could have varying degrees of impact on housing prices at different locations. This global-scale spatially varying relationship is known as continuous spatial heterogeneity in the housing market. Furthermore, discontinuous spatial heterogeneity of the housing market can occur at local spatial scales within an urban area with complex structures. For example, the influence of traffic accessibility on housing prices may change gradually from the city center to the suburbs as the distance from the residence location to the city center increases. Meanwhile, significant price changes may occur in the surrounding local areas of district boundaries, famous streets, or school districts. Therefore, identifying multi-scale spatial heterogeneity is favorable to understand the characteristics of the urban housing market and avoid any misspecification. This study proposes a fusion model that combines the random effects eigenvector spatially filtering-based spatially varying coefficient (RE-ESF-SVC) model with fused lasso to extract spatial heterogeneity at different spatial scales simultaneously. We apply the proposed method to residential rent data in the Tokyo metropolitan area. It successfully detects continuous and discontinuous spatial heterogeneity in the rental market associated with different housing attributes and locational determinants. This study provides new insights to explore how related influencing factors determine and characterize housing prices in urban areas.
Identifying multi-scale spatial heterogeneity in the urban housing market: A fusion approach combining random effects eigenvector spatially filtering-based spatially varying coefficient (RE-ESF-SVC) model with fused lasso
Category
Virtual Paper Abstract
Description
This abstract is part of a session. Click here to view the session.
| Slides