Interarea indexes of the cost of shelter using hedonic quality adjustment techniques

The cost of shelter is the single most important component of interarea differences in the cost-of-living. This paper constructs hedonic interarea indexes of the cost of shelter using a major new source of detailed microdata, the CPI Housing Survey. Identification of the exact location of housing units permits us to control for measured neighborhood characteristics, and variance components models are used to model the effects of unmeasured neighborhood characteristics. Hierarchical models with interarea slope coefficient variation are applied and model performance is evaluated by comparing out-of-sample predictions. The hierarchical approach allows us to pool information from different cities to reduce the variability of the areasʹ coefficients. A multilateral index number formula is employed in the index-number construction.