The dynamic correlation and volatility of real estate price and rental: an application of MSV model

Abstract

This article investigates the dynamic conditional correlation between price and rental among four types of real estate markets, namely housing, retail, office and factory. In the case of Hong Kong (from January 1993 to December 2008), the Dynamic Conditional Correlation Multivariate Stochastic Volatility (DCC–MSV) model and Bayesian Markov Chain Monte Carlo (MCMC) are employed to capture the volatility and time varying correlation of property price and rental in each market. The empirical results demonstrate that the volatilities of price are significantly larger than those of rental for each type of real estate markets throughout the whole sample period. Besides, the findings indicate that the correlations between price and rental are time-varying, and the average correlation in the housing market is much larger than those in the other three markets. Also, the conditional correlations in the factory market are the most volatile, while those in the office market are relatively stable. Additionally, two particularly volatile periods have been identified in and around 1997 and 2003.