Data Sources for Measuring House Price Changes

Abstract

The effectiveness of house price index methods depends on the quality and appropriateness of the data employed. This article describes and compares the types of data available. The comparison criteria include ability to adjust for quality; availability of buyer and seller characteristics, financing information, length of time on the market, and locational detail; representativeness; sample size; availability lag; length of time covered; and cost. The results highlight the importance of the match between data type and the proposed use of the house price index. While no data type is ideal, transaction and assessment data are highly rated because of their completeness and increasingly widespread availability. Mortgage transaction data are valuable for evaluating pools of mortgages and modeling defaults. Multiple listing service data are often well suited to local studies. The American Housing Survey has the most detailed property characteristics but must be fairly highly aggregated because of sample-size limitations.