Search results for: multicollinerity.

Authors: Yan Wang, Yasushi Asami, Yukio Sadahiro

Abstract:

In urban area, several landmarks may affect housing price and rents, and hedonic analysis should employ distance variables corresponding to each landmarks. Unfortunately, the effects of distances to landmarks on housing prices are generally not consistent with the true price. These distance variables may cause magnitude error in regression, pointing a problem of spatial multicollinearity. In this paper, we provided some approaches for getting the samples with less bias and method on locating the specific sampling area to avoid the multicollinerity problem in two specific landmarks case.

Keywords: Landmarks, hedonic regression, distance variables, collinearity, multicollinerity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1905

Authors: Yan Wang, Yasushi Asami, Yukio Sadahiro

Abstract:

In urban context, urban nodes such as amenity or hazard will certainly affect house price, while classic hedonic analysis will employ distance variables measured from each urban nodes. However, effects from distances to facilities on house prices generally do not represent the true price of the property. Distance variables measured on the same surface are suffering a problem called multicollinearity, which is usually presented as magnitude variance and mean value in regression, errors caused by instability. In this paper, we provided a theoretical framework to identify and gather the data with less bias, and also provided specific sampling method on locating the sample region to avoid the spatial multicollinerity problem in three distance variable’s case.

Keywords: Hedonic regression, urban node, distance variables, multicollinerity, collinearity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1994