The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Articles | Volume XLII-2/W2
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W2, 123–125, 2016
https://doi.org/10.5194/isprs-archives-XLII-2-W2-123-2016
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W2, 123–125, 2016
https://doi.org/10.5194/isprs-archives-XLII-2-W2-123-2016

  06 Oct 2016

06 Oct 2016

3D WEIGHT MATRICES IN MODELING REAL ESTATE PRICES

A. Mimis A. Mimis
  • Dept. of Economic & Regional Development, Panteion University, Greece

Keywords: Spatial Econometrics, 3D, Spatial Analysis, Weight Matrices

Abstract. Central role in spatial econometric models of real estate data has the definition of the weight matrix by which we capture the spatial dependence between the observations. The weight matrices presented in literature so far, treats space in a two dimensional manner leaving out the effect of the third dimension or in our case the difference in height where the property resides. To overcome this, we propose a new definition of the weight matrix including the third dimensional effect by using the Hadamard product. The results illustrated that the level effect can be absorbed into the new weight matrix.