Spatial cluster detection aims at detecting a particular region in which the hypothesis of random occurrence of an event is rejected. This information is of extreme relevance in epidemiology studies. This article presents a cluster detection method that aggregates the spatial neighbor structure into the cluster growing process. The procedure allows the detection of arbitrarily shaped clusters. Standard scan spatial statistics confine the cluster geometry shape to circular shaped clusters, resulting in partial or over sized clusters. Restrictions during the growth process are suggested in order to prevent over sized clusters with odd geometries. An evaluation of the method's performance is provided through simulation studies. Results of cluster detection in crime data are presented for Belo Horizonte city. Referencias Bibliograficas [ACT+06] Assuncao, R., Costa, M., Tavares, A., Ferreira, S. (2006), Fast detection of arbitrary shaped disease clusters. Statistics in Medicine. Forthcoming [BN91] Besag, J. and Newell, J. (1991), The detection of clusters in rare diseases. Journal of the Royal Statistic Society A, vol. 154, pages 143-155. [CA05] Costa, M. A., Assuncao, R. M. (2005), A fair comparison between the spatial scan and the BesagNewell disease clustering tests. Environmental and Ecological Statistics, vol. 12, pages 297-315. [DA04] Duczmal, L. and Assuncao, R. (2004), A simulated annealing strategy for the detection or arbitrarily shaped spatial clusters. Computational Statistics & Data Analysis, vol. 45, pages 269-286. [KTP03] Kulldorff, M., Tango, T., Park, P. J., (2003), Power comparisons for disease clustering tests. Computational Statistics & Data Analysis, vol. 42, pages 665-684. [Kull97] Kulldorff, M. (1997), A Spatial Scan Statistics. Commun. Statist. Theory and Methods., vol. 26, pages 1481-1496. [OCC+88] Openshaw, S., Craft, A. W., Charlton, M. and Birch, J. M. (1988), Investigation of leukaemia clusters by use of a geographical analysis machine. Lancet i, pages 272-273 [TT05] Tango, T., Takahashi, K. (2005), A flexibly shaped spatial scan statistic for detecting clusters. International Journal of Health Geographics, 4:11. Agradecimentos Os autores agradecem a FAPEMIG e a Pro-Reitoria de Pesquisa da UFMG pelo apoio financeiro e a Policia Militar de Belo Horizonte pelo auxilio e suporte a base de dados.
[1]
Julian Besag,et al.
The Detection of Clusters in Rare Diseases
,
1991
.
[2]
Peter J. Park,et al.
Power comparisons for disease clustering tests
,
2003,
Comput. Stat. Data Anal..
[3]
T. Tango,et al.
International Journal of Health Geographics a Flexibly Shaped Spatial Scan Statistic for Detecting Clusters
,
2005
.
[4]
Marcelo Azevedo Costa,et al.
A fair comparison between the spatial scan and the Besag–Newell Disease clustering tests
,
2005,
Environmental and Ecological Statistics.
[5]
M. Kulldorff.
A spatial scan statistic
,
1997
.
[6]
Renato Assunção,et al.
A Simulated Annealing Strategy for the Detection of Arbitrarily Shaped Spatial Clusters
,
2022
.
[7]
R. Assunção,et al.
Fast detection of arbitrarily shaped disease clusters
,
2006,
Statistics in medicine.
[8]
A. Craft,et al.
INVESTIGATION OF LEUKAEMIA CLUSTERS BY USE OF A GEOGRAPHICAL ANALYSIS MACHINE
,
1988,
The Lancet.