Considering the importance of housing sector in national economy, it is natural that many researchers are interested in the accurate prediction of housing market indicators. They have employed time series models such as the vector autoregressive(VAR) ...
Considering the importance of housing sector in national economy, it is natural that many researchers are interested in the accurate prediction of housing market indicators. They have employed time series models such as the vector autoregressive(VAR) model. However, as an analytical tool, VAR has weakness points that require estimation of large number of parameters and is sensitive to the researcher’s discretion. This paper experiments the Bayesian VAR, which requires a fewer number of parameters, as an alternative in the prediction of future changes of house price index. We find evidence to support that the Bayesian VAR shows better performance than VAR. The Bayesian VAR could be applied more broadly in the prediction of future economic trends.