Property valuation and the structure of urban housing markets

摘要: Since the 1980s UK academics have promoted use of multiple regression analysis in property valuation. Recently, however, there has been growing recognition that models will be subject to aggregation bias if they fail accommodate existence housing market segmentation (submarkets). In this study, we compare empirical performance a standard hedonic house price model for city Glasgow with segmented which recognises importance understanding underlying structure and, particular, submarkets different dwelling types. The results show (weighted) error is significantly lower than wide model. Consequently, propose two‐stage approach application MRA techniques residential First, following traditional institutional markets, should subdivided into distinct structurally differentiated segments. These segments can usefully identified by principal components factor allows identification most important common bundle. Second, separate equations estimated each segment. Although best‐fit equation may vary from sector likely reflect behavioural realities market, and provide basis more accurate valuations.