In this paper we present the repeat sales index methodology developed by Case and Shiller (1987) and its estimation problem. We particularly describe the problem arising from the time intervals construction for the estimation. We then apply this methodology to the Paris residential market. We use the CD-BIEN database that contains more than 220 000 repeat sales transactions for residential properties in the Paris area covering the period 1973-2001 period. This index based on returns is compared to the official one used in France for Paris based on single prices, the Notaires/INSEE index. We then underline the robustness of in the index estimation according to its periodicity by the way of the return and volatility estimation. The index sensibility to the time period is studied in the last part. We conclude that i) the estimation is quite robust whatever the estimation period is, and ii) this index is significantly different from the official residential index for Paris.
BARONI, M., BARTHELEMY, F. et MAHDI, M. (2004). Physical Real Estate: A Paris Repeat Sales Residential Index. ESSEC Business School.