Semi-parametric Cluster Detection
A semi-parametric density ratio testing method which borrows strength from two or more samples is applied to moving windows of variable size in cluster detection. The method requires neither the prior knowledge of the underlying distribution nor the number of cases before scanning. A Monte Carlo power study shows that given a cluster candidate, under certain conditions the semi-parametric density ratio method achieves a relatively higher power than the power achieved by Kulldorff‘s celebrated scan statistics method and by a certain focused test in testing the hypothesis of no cluster. The semi-parametric method potential in cluster detection is illustrated using both simulated and real spatial data.
AMS Subject Classification62F03
KeywordsCombined data exponential tilt likelihood moving window power scan statistics
Unable to display preview. Download preview PDF.
- Fokianos, K., 2006. Density Ratio Model Selection. Journal of Statistical Computations and Simulation. To appear.Google Scholar
- Gagnon, R., 2005. Certain Computational Aspects of Power Efficiency and of State Space Models. Ph.D Dissertation, Department of Mathematics, University of Maryland, College Park, Maryland.Google Scholar
- Pickle, L.W., Feuer, E.J., Edwards, B.K., 2003. U.S. Predicted Cancer Incidence, 1999: Complete Maps by County and State From Spatial Projection Models, National Cancer Institute, Cancer Surveillance Monograph No. 5, NIH Publication No. 03-5435.Google Scholar