The purpose of this paper is to address the heterogeneity of real estate assets with regard to investment risk measurement, with Paris’ apartment market as a case study. Quantile regression is used to handle the fact that willingness to pay for housing attributes may vary greatly over both space and asset value categories. The method is alternately applied on central and peripheral districts of Paris, or “arrondissements”, with hedonic indices built for nine deciles over a 17-year period (1990-2006). Portfolio allocation is subsequently analysed with deciles being the assets.The findings suggest that during the slump, peripheral districts show better resilience and define the efficient frontier while also exhibiting a lower volatility. In addition, higher returns are observed for lower-priced apartments, both central and peripheral. During the recovery and boom stages of the cycle, the highest returns are experienced for the cheapest apartments in central locations, whereas upper-priced, centrally located units yield the lowest returns. The originality of this research resides in the application of quantile regression in a real estate investment and risk management context. The methodology may raise individual investors’ and practitioners’ attention, especially index providers’. Link to the article
AMÉDÉE-MANESME, C.O., BARONI, M., BARTHELEMY, F. and DES ROSIERS, F. (2017). Market Heterogeneity, Investment Risk and Portfolio Allocation: Applying Quantile Regression to the Paris Apartment Market. International Journal of Housing Markets and Analysis, 10(5), pp. 641-661.