Can street segments indexed for accessibility form the basis for housing submarket delineation?. Issue 7 (2nd October 2016)

• Record Type: Journal Article
• Authors: Xiao, Yang; Webster, Chris; Orford, Scott
• Is Part Of: Housing studies. Volume 31:Issue 7(2016)
• Page Start: 829
• Page End: 851
• Publication Date: 2016-10-02
• Subjects: Housing submarket -- street segments -- accessibility -- geographic boundary -- network analysis; Housing -- Periodicals; 307.336
• DOI: 10.1080/02673037.2016.1150433
• Languages: English
• ISSNs: 0267-3037

Abstract

Abstract: We test an approach to spatial housing submarket delineation using street segment as the spatial unit and using finely grained measures of accessibility derived from spatial network analysis. The underlying idea is that street segment connectivity captures fine variations in homebuyers' preferences for the location. The advantage of the approach is that it is spatially fine grained; it uses the street segment, intuitively the most fundamental spatial unit for spatial housing market analysis; it allows the use of statistical tests to optimize within-submarket similarities, identifying spatial groups of street segments with the most similar accessibility features; it avoids the predefined arbitrary geographic boundaries usually used in spatial submarket delineation; it increases the variability of accessibility information in submarket delineation, accessibility being the principal spatial determinant of housing price; and it allows for normalized measures of accessibility at different spatial scales making it appropriate for comparative analysis across cities and across time. Using a case study of Cardiff, UK, we compare the results with a market segmentation scheme based on prior-knowledge, notably one relying on building-type classification. We conclude that street layout can be used to efficiently delineate housing submarkets, and that the estimation is very close to the scheme requiring prior-knowledge. It has advantages, however, that make it worthy of further investigation, namely its adaptability, scale-specificity and lower reliance on local knowledge of housing market culture and data.