460 – Forecasting and Official Statistics
House Price Tiers in Repeat Sales Estimation
Douglas McManus
Freddie Mac
The existing methods for estimating tiered house price indexes are subject to substantial biases because a property's tier can only be measured imprecisely. Both academic research and industry practice have implemented tiered house price index estimation techniques but without a methodological solution to the bias problem. This paper proposes bootstrap procedures for correcting this bias in the context of testing for the existence of house price tiers. This method is illustrated at the state level for California, and it is shown that there are statistically significant tier effects.