Housing market cycles in large urban areas

Abstract

This paper examines the behaviour of house prices in large metropolitan areas. Using a sample of metropolises it is shown that real estate prices are largely nonlinear. It is found that dynamic asymmetries in the housing market cycle can well be modelled using a logistic smooth transition model (LSTAR). Further, it is found that the LSTAR specification has better forecasting properties with respect to other linear and nonlinear models.