José Maria Montero Lorenzo, Ph.D., University of Castilla-La Mancha, Cobertizo de San Pedro Martir, s/n,, Toledo, 45071, Spain, Gema Fernández-Avilés Calderón, Associated, Teac, Castilla-La Mancha University, Cobertizo de San Pedro Mártir s/n, Toledo, 45071, Spain, Jorge Mateu, Ph.D., University Jaume I of Castellon, Campus Riu Sec, Castellon, 12071, Spain, and Emilio Porcu, M.B.A., Departamento de Matemáticas, University Jaume I, Avenida Vicente Sos Bainat S/N, Castellón, 12071, Spain.
1. Title: “Making Environmental Quality Indexes for Real Estate Valuation”
2. Objectives: The problem that usually arises when environmental information is included in hedonic property price models is that here is a mismatch between the spatial ‘support’ for the environmental measured variables and the property prices. In the specialized literature, the usual solution to this problem is to interpolate the environmental variable(s) to obtain their interpolated values in the locations where property prices are available. It can be considered three possibilities: (i) interpolate (preferably cokriging) such variables and include all variables in the model; (ii) elaborate an environmental index, and then interpolating it (preferably kriging); and (iii) interpolate (preferably cokriging) the environmental variables considered and, subsequently, elaborate an environmental index.
3. Data/Methods: In this paper, options (ii) and (iii) are empirically compared using six environmental variables to elaborate an Environmental Quality Index in Madrid City (Spain).
4. Results/Expected results: Option (i) is preferred when dealing with only one environmental variable, because the inclusion of several variables in a hedonic property price model is not an easy task. If several environmental variables are included in the analysis, option (ii) is the one chosen in the specialized literature, arguing that it is a way to transform a multivariate problem in a univariate one. The last statement being true, our proposal is option (iii) because the variance of the estimation errors is lesser than using (ii).
5. Conclusion: This case study empirically confirms an important aspect of the geostatistical theory when dealing with several variables and the objective is to transform a multivariate problem in a univariate one: In general, prediction MSE is lesser when interpolating the variables involved in a linear combination and then elaborating such a linear combination.
6. Key words: environmental index, interpolation, kriging, hedonic property price models.