A Geo-Statistical Method to Define Districts within a City
Posted: 11 Mar 2003
Abstract
The aim of this study is to present a methodology that may be used for districting, that is, to divide a city into homogeneous districts, using the housing characteristics of the city as the criterion for such a division. The methodology is based on the use of Principal Component Analysis into the Theory of Regionalized Variables. Moreover, it included a series of multivariate techniques to verify the degree of discrimination achieved between the various districts identified. The theoretical description is complemented by the application of this districting methodology to the city of Granada. Furthermore, the districting obtained throughout this methodology is compared with that established by the public administration. This verifies which of the two districtings achieves that the defined districts be different, as regards the housing characteristics.
Keywords: Theory of Regionalized Variables, Block Kriging, Iterative Residual Kriging, Principal Components Analysis, Canonical Discriminant Function Analysis, Districting
Suggested Citation: Suggested Citation