The Estimation of Spatial Effects of the Office Rent in Seoul

This study analyzed the spatial effects on office rent in three representative regions in Seoul (CBD, Gangnam, and Mapo/Ysouido) using three autocorrelation models (SEM, SLM, and SAC) as well as the hedonic model (traditional OLS). The results of this study show that SEM is more suitable than the typical regression model (OLS) for office market analysis. Moran's I coefficient which can presume the spatial effects represented the spatial dependence. The strongest spatial dependence was observed in CBD, followed by Mapo/Yeouido and Gangnam. It is caused by the difference between the rent level and the spatial density rate. But this spatial effects of office market in Seoul were weak as compared with that of housing market because of supplier's market that the rent is decided by suppliers of office, so far. Despite the drawback that the analysis is unable to control the effects of omitted-variables completely, it is a meaningful study dealing with the spatial effects of office market in Seoul.