Uniformly consistent tests for contamination neighbourhoods

H. Inouzhe Valdes

Often in statistical practice a model offers a good description of
the data even though it is not the 'true' random generator. A classical test
of fit would reject such a model if the sample is large enough. We consider a
more flexible approach based on contamination neighbourhoods around a model.
Using trimming methods and the Kolmogorov metric we introduce tests of fit
to some contamination neighbourhoods which have uniformly exponentially small
type I and type II error probabilities. As as application we explore a
credibility analysis of descriptive models.

Palabras clave: trimming Kolmogorov distance contamination neighbourhoods uniformly exponential hypothesis testing

M. C. Martín Blanco, A. Jiménez Martín, A. Mateos Caballero

A. Shatla, N. Corral Blanco, C. E. Carleos Artime