A statistical method for the identification of spatial clusters.
Using data from pest reports in the city of Newcastle upon Tyne, England, the writer describes a way of identifying and locating clusters in data reported in small areal units. He aggregates areal units into zones that are not prespecified at the beginning and that can be of any size or shape. He...
| Publicado en: | Environment & Planning A Vol. 29; pp. 1039 - 1056 |
|---|---|
| Autor principal: | |
| Formato: | Artículo |
| Publicado: |
Pion Limited
June 1997
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| Sumario: | Using data from pest reports in the city of Newcastle upon Tyne, England, the writer describes a way of identifying and locating clusters in data reported in small areal units. He aggregates areal units into zones that are not prespecified at the beginning and that can be of any size or shape. He outlines a statistical distribution against which any apparent clusters in the zones can be examined for statistical significance. He uses Monte Carlo methods to derive the distribution, and he contends that this method can be used to represent complicated null hypotheses. He asserts that in this approach the vital distinction between prespecified and constructed areas or zones of interest is recognized and is shown to avoid some of the problems generally associated with modifiable areal units. |
|---|