A VARYING PARAMETERS APPROACH TO CONSTRUCTING HOUSE PRICE INDEXES

Conventional housing price index models assume interperiod parameter s tability and typically employ either repeat sales or hedonic methodolo gies. This paper introduces a method of index construction that combin es multiple sales observations with single sale transactions while per mitting characteristics prices from hedonic regressions to vary over t ime. A test for interperiod parameter stability is provided. Each peri od's data are arranged by location and repeat sales are matched by row s. This construction allows greater use of sample information and ackn owledges the unique contribution of repeat sales to the estimation pro cess. It also produces intertemporal error correlations that can be be neficially exploited by the seemingly unrelated regressions (SUR) tech nique. The paper also demonstrates a significance test for error corre lation and discusses the treatment of unequal numbers of observations among index periods.