VARIANCE DECOMPOSITION OF REAL ESTATE RETURNS

Nominal property returns can be broken down into two components: real property returns and inflation. A structural VAR model is proposed to estimate the relationships between nominal property return, real property returns and inflation. Bayesian Methods (Markov Chain Monte Carlo) are used to estimate the system. Non Information Normal Wishart priors and the Gibbs sampling algorithm are used to carry out the posterior simulations. Economic theory is used to impose restrictions on the system. The results are helpful in distinguishing which factors account for higher variance in property returns.