The Land Market in Hong Kong : Price Index and Relationship with the Property Market

Housing market is highly heterogeneous and is most accurately represented as a collection of distinct yet related submarkets. In this paper, a flexible semiparametric model is developed to model housing submarkets. The k Nearest Neighbor estimates of hedonic price coefficients, varying both spatially and temporally, are used to identify housing submarkets by Classification and Regression Tree method. The proposed semiparametric model is applied to identify the submarkets in the Hong Kong residential property market. The heterogeneity of local housing market is illustrated by comparing hedonic price coefficients and housing traits of flats within each housing submarket