"In this paper, the similarities and differences in housing charateristics among the different ìarrondissementsî of Paris, France, are brought out using a twofold, spatio-temporal approach: while standard hedonic price modelling is applied to a series of building and neighbourhood attributes in order to define spatial submarkets, differentials in price appreciation over time are also measured using both hedonic and repeat sales indices.Spatial segmentation is performed on the 20 Paris ìarrondissementsî as well as on the 80 neighbourhoods, or ìquartiersî (each ìarrondissementî is composed of four ìquartiersî), with cluster analysis being resorted to for that purpose. The database, which is provided by the Chambre des Notaires de France, includes cases spread over a 17 year period, that is from 1990 to 2006. Housing descriptors include, among other things, the building age, the apartment size and the number of rooms, the presence of a garage, the type of street and access to building (boulevard, square, alley, etc. as well as a series of location dummy variables standing for the ìarrondissementsî and ìquartiersî.""