Model House Price Volatilities: Spatial and Temporal Structure

Abstract

We propose a new type of house price model for risk management and asset pricing, different from the prevailing time series type of models commonly used by the academia and industry. It separates the modeling of HPA expectation and volatility. It models the essential features of house price volatilities in the simplest form. In terms of the volatility term structure, it models the house price mean reversion and momentum and exhibits short-term positive serial correlation and long-term negative serial correlation. It models the spatial correlation as well as local features of volatility term structure. It is in continuous time form, has only three parameters, and is analytically tractable. The model estimation is transparent, simple, robust, and accurate.