This software provides means to perform exploratory statistical analysis for the non-temporal spatial data. The goal is to deduce conditions associated with the level of the response in different areas of the studied region. The data should be compiled for the census areas, defined by the regional administrative division. Additionally, the tool allows estimating accuracy of the approximation by a statistical model to the higher resolution data, defined on the GIS lattice. Software is supplied in several packages.
How the tool works
The regional data is summarized by a multiple regression model with the unknown joint distribution of error terms. Hypothesis testing for model evaluation is done by the permutations method. Confidence intervals for regression estimates are constructed using bootstrapping approach. Accuracy of model approximation to the GIS lattice is assessed in every cell and summarized by statistics from the derived sample of accuracy values. Additionally, approximation accuracy can be estimated for a cluster of cells in the GIS lattice.
Downloads
Packages: lu-preprocessing-1.0.0-standalone.jar
lu-rescaling-1.0.0-standalone.jar
lu-regression-1.0.0-standalone.jar
lu-approximation-1.0.0-standalone.jar
regression-tests-1.0.0-standalone.jar
Version: 1.0.0
Source code: code.zip
Example data: example data.zip
Manuals: Introduction to Exploratory Spatial Analysis
Permutation Hypothesis Testing and Bootstrapping
Testing Spatial Autocorrelation with the Bootstrap
Acknowledgements
This work received support from the EU FP7 project COMPLEX (grant no. 308601)
PUBLICATIONS
COMPLEX PROJECT
Knowledge Based Climate Mitigation Systems for a Low Carbon Economy
Related links
International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313