The repeat-sales methodology has become a standard approach for estima
ting real estate price indices. This article examines the underlying a
ssumptions inherent in the repeat sales model and provides an empirica
l test for both included and omitted variables as sources of aggregati
on bias. The results indicate that virtually all price indices may be
biased, the degree of bias being dependent upon the number of variable
s examined and the instability of their parameters over time.