zbMATH — the first resource for mathematics

The power of focused score tests under misspecified cluster models. (English) Zbl 0964.62104
Lawson, Andrew (ed.) et al., Disease mapping and risk assessment for public health. Chichester: Wiley. 257-269 (1999).
From the introduction: In recent years there has been considerable interest in the spatial distribution of disease cases and possible relationships to environmental exposures. Of particular interest are possible temporal or spatial aggregations of incident cases, i.e. ‘disease clusters’, defined as ‘a bounded group of occurrences related to each other through some social or biological mechanism, or having a common relationship with some other event or circumstance’.
Score tests of focused clustering seek to maximise statistical power versus local alternatives. A specified local alternative, corresponding to an a priori model of focused disease clustering, defines the particular form of the test. We explore the power of misspecified score tests, i.e. tests defined for one type of cluster model applied to clustering generated from a different model.
For the entire collection see [Zbl 0942.00010].
62P10 Applications of statistics to biology and medical sciences; meta analysis