Abstract
This paper analyzes the spatio-temporal interaction patterns of the house sales and rent price using SpVAR model and evaluate accuracy of estimation and prediction of suggested model, compared with panel VAR model. For the case study, we analyzed the house sales and rent price by housing types for 65 sigungu of the Seoul metropolitan area. The result shows that the 1-period former neighboring rental price of apartment has effects on all types of rental prices, whereas the sale price of apartment has little influences. Especially, the increase of apartment rental prices for a particular area substantially impacts multi-family housing rental prices for neighboring areas. This result empirically supports the spillover effect of the house sales and rent price. Also, the SpVAR model predicted the direction of the house price variation more accurately than the non-spatial VAR model. This means that the spatial association in house prices by housing type plays an important role in determining sub-market house prices, and the house price of a particular housing type influences other house prices in neighboring areas.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
