Spatially disaggregated real estate indices

A spatial-temporal Markov random-field model is used to produce indices for residential real estate from repeat home sale data. A set of regions is re presented by a graph in which neighboring regions are linked. This graph. r epeated consecutively a number of times, with each region linked to the sam e region at adjacent times, defines a spatial-temporal graph that connects regions over space and time in which each node represents a region at a par ticular time. An index is defined at each node to be the rate of appreciati on of log sale price in a region during the preceding time interval. The in dices are estimated from data consisting of repeat home sales. The Markov r andom-field model specifies spatial and temporal correlations between neigh boring indices and relations between indices and individual repeat sales. A method is proposed for estimating various parameters in the model and for obtaining real-estate indices. Following this prescription. indices are cal culated for the Dade County, Florida, residential real-estate market. Resul ts are compared to those obtained using competing space-only models.