이분성 발견 및 미발견시의 주택하위시장 구분방법

The Precise Distinction of Homogeneous Housing Submarkets With and Without Heteroskedasticity- Focused on the Concrete Methodology -

When it comes to the market segmentation in the housing market, the aggregate models, i.e., the distinction of homogeneous housing submarkets, are only meaningful if the underlying phenomenon is homogeneous across the units of observation. Unless there is a homogeneous spatial process underlying the data, any aggregation will tend to be misleading. If this is not the case, both the heterogeneity and structural instability should be acconted for in an aggregation scheme. This aspect of modifiable area unit problem should be considered as a specification issue to the form of spatial heterogeneity, and not solely as an issue determined by the spatial organization of the data. For situations where one is trying to uncover the difference the stability of parameter between the housing submarkets, one is left with the question how to proceed. Then the Chow statistic will apear as the one of the various methods to tackle it on the assumptic homoskedasticity. Especially when it finds the with-in heteroskedasticy, the usage of critical values from the F-distribution is suspectecd, even if the average variance is the same across samples. The Wald test is more general and obviously applies under conditions of heteroskedasticity. Therefore, this paper focuse on the precise distinction of homogeneous housing submarkets with and without heteroskedasticity.