Higher Frequency Hedonic Property Price Indices: A State Space Approach

The hedonic imputation method estimates separate characteristic shadow prices for each period. These are used to construct matched samples, which are inserted into standard price index formulas. We implement two innovations to improve the method’s effectiveness on housing data at higher frequencies. First, we use a time-varying parameter model in state-space form to increase the reliability of the estimated characteristic shadow prices. Second, we significantly reduce the number of parameters by replacing postcode dummies by a geospatial spline surface. Empirically, using a novel criterion, we show that in higher frequency comparisons our hedonic method outperforms competing alternatives.