European Regional Science Association
|
European Regional Science Association
|
|
|
|
|
The abstract for paper number 247:
Juan Carlos Duque Cardona, AQR Research Group. Universitat de Barcelona, Barcelona, Spain, Jordi Surinach, AQR Research Group
Universitat de Barcelona, Barcelona, Spain
Automated Regionalisation: Methodological and Empirical Issues
One of the main questions to solve when analysing geographically added information consists of the design of territorial units adjusted to the objectives of the study. This is related with the reduction of the effects of the Modifiable Areal Unit Problem (MAUP).
In this paper an optimisation model to solve regionalisation problems is proposed. This model seeks to reduce some disadvantages found in previous works about automated regionalisation tools. Between its main characteristics are: a) The possibility of solve objectives of segmentation or comparability, b) Incorporation of own characteristics of the elements and its relations, c) the possibility of initiating the regionalisation process without determining non-optimal solutions as departure point, and d) the conformation of regions that fulfill the geographic continuity with a greater degree of freedom in forms that can adopt these regions.
In the last part of the paper, the model is applied to the 38 statistics zones in Barcelona (Spain). The results show that, using a controlled regionalisation, the statistical inference for each conformed zone is significantly more fit.
Key words: Regionalisation, Modifiable Areal Unit Problem, Optimisation, Contiguity constraint.
Unfortunately full paper has not been submitted.