Temporal and Spatial Information Diffusion in Real Estate Price Changes and Variances

This article examines patterns of temporal and spatial diffusion of real estate price changes. In addition to means, changes in volatility are tracked in reaction to substantial new information, estimated with GARCH-M methods. The data covers towns in Connecticut and near San Francisco. There is evidence of negative feedback at short lags, contrary to previous research on housing and other assets. There is also evidence of a moving average error process which tends to reverse recent shocks. Significantly positive spatial information diffusion is found from neighboring towns in Connecticut but none in control tests on non-neighboring towns. The results also include evidence of a risk-reward tradeoff in housing price changes in the San Francisco area