Heterogeneity in spatial correlation and influential factors on property prices of submarkets categorized by urban dwelling spaces

The relationship between housing characteristics and residential property price is complex to illustrate. To have a better understanding of the housing-price structure, structural or spatial submarkets have been introduced in previous studies. However, there is no consensus on the importance of neighborhood effect among them. With regard to this issue, this study specifically focuses on the differences in structural, geographical, and demographical effects between housing submarkets (as categorized by dwelling space). Market subdivisions are relative to consumer behaviors. The heterogeneity of spatial effects in different submarkets gives an explanation as to why the impact of nonneighborhood housing factors in those submarket models is similar, regardless of whether the neighborhood effect is included or not. It also indicates that only some, but not all, of the potential purchasers would make decisions with reference to transaction records of flats in the surrounding areas, only when the flat they want to buy is within a specific range in size. Other critical characteristics such as building age, dwelling space, location choice, and commuting time (to central business district) show remarkably different effects on housing price in those submarkets. Segmentation of housing market by dwelling space (and the average price per square feet is higher when the flats are larger) helps to obtain a better understanding of the importance of dwelling space in housing valuations and in urban housing planning.