본문 바로가기
  • Home

Exploring Spatial Heterogeneity in Factors for Apartment Prices in Seoul Using Moran Eigenvector Spatial Filtering Based on Spatially Varying Coefficients

  • Journal of the Association of Korean Geographers
  • Abbr : JAKG
  • 2019, 8(2), pp.321-335
  • DOI : 10.25202/JAKG.8.2.14
  • Publisher : Association of Korean Geographers
  • Research Area : Social Science > Geography
  • Received : June 27, 2019
  • Accepted : July 31, 2019
  • Published : August 31, 2019

Hyeongmo Koo 1

1中國南京師範大學 地理科學學院

Accredited

ABSTRACT

Spatially varying coefficients models have been used as a tool for exploring spatial heterogeneity in regression coefficients. But, among various spatially varying coefficients models, only a geographically weighted regression (GWR) has been used for exploring spatial heterogeneity in factors for housing prices although it suffers from its low explanatory power and high multicollinearity between local coefficients. This study applies another type of a spatially varying coefficients model based on Moran eigenvector spatial filtering (MESF) to explore factors to apartment sale prices. The analysis results are compared to those of GWR and global MESF. The result shows MESF outperforms GWR in terms of explanatory power, spatial autocorrelation in model residuals, and multicollinearity between local coefficients. In additions, a variation in local coefficients against their global coefficients suggests that exploring spatial heterogeneity is necessary for a housing price estimation model.

Citation status

* References for papers published after 2022 are currently being built.