The choice of methodology for computing housing price indexes: Comparisonsof temporal aggregation and sample definition

Housing transactions are executed and recorded daily, but are routinely poo led into longer time periods for the measurement and analysis of housing pr ice trends. We utilize an unusually rich data set, covering essentially all arm's length housing sales in Sweden for a dozen years, in an attempt to u nderstand the effect of temporal aggregation upon estimates of housing pric es and their volatilities. This rich data set also provides a unique opport unity to compare the results using the conventional weighted repeat sales m odel (WRS) to those based on a research strategy which incorporates all ava ilable information on house sales. The results indicate the clear importanc e of temporal disaggregation in the estimation of housing prices and volati lities-regardless of the model employed. The appropriately disaggregated model is then used as a benchmark to compar e estimates of the course of housing prices produced by the two models duri ng the twelve year period 1981-1993. These results indicate that much of th e difference between estimates of price movements can be attributed to the data limitations which are inherent in the repeat sales approach. The resul ts, thus, suggest caution in the interpretation of government-produced pric e indices or those produced by private firms based on the repeated sales mo del.