Modeling The Choice Of Dwelling Unit In A Condominium Complex

This paper studies dwelling unit choice models in a condominium complex to en-hance knowledge about the willingness to pay for dwelling unit attributes, which can quantify the significance of the influence of dwelling unit attributes on housing choice decision and provides real estate developers a reference for pricing the dwelling unit. Condominium living is the most common residential choice for citizens in China. To the best of the author’s knowledge, this paper is the first of its kind to investigate the choice of dwelling units within a condominium complex. By using revealed preference data from a condominium complex in Beijing, we extract a variety of dwelling unit at-tributes, including building location, unit location, floor, orientation, size, price, etc. The multinomial logit model, mixed logit model and latent class choice model are estimated, in which the hierarchical Bayes method is employed to estimate the coefficients of the mixed logit model. The multinomial logit model and the mixed logit model can get very similar willingness to pay estimates, but the mixed logit model reveals strong evidence that considerable behavioral heterogeneity exists in dwelling unit choice and the substitu-tion patterns among dwelling units are fairly complex. The latent class choice model ac-commodates the taste variations problem through a semiparametric approach and shows that the home buyers can be divided into two segments, which pay different attentions to certain attributes, including building far from the main road, end unit and highest floor.