On combinatorial testing problems
Addario-Berry, Louigi ; Broutin, Nicolas ; Devroye, Luc ; Lugosi, Gábor
Ann. Statist., Tome 38 (2010) no. 1, p. 3063-3092 / Harvested from Project Euclid
We study a class of hypothesis testing problems in which, upon observing the realization of an n-dimensional Gaussian vector, one has to decide whether the vector was drawn from a standard normal distribution or, alternatively, whether there is a subset of the components belonging to a certain given class of sets whose elements have been “contaminated,” that is, have a mean different from zero. We establish some general conditions under which testing is possible and others under which testing is hopeless with a small risk. The combinatorial and geometric structure of the class of sets is shown to play a crucial role. The bounds are illustrated on various examples.
Publié le : 2010-10-15
Classification:  Hypothesis testing,  multiple hypotheses,  Gaussian processes,  62F03,  62F05
@article{1283175989,
     author = {Addario-Berry, Louigi and Broutin, Nicolas and Devroye, Luc and Lugosi, G\'abor},
     title = {On combinatorial testing problems},
     journal = {Ann. Statist.},
     volume = {38},
     number = {1},
     year = {2010},
     pages = { 3063-3092},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1283175989}
}
Addario-Berry, Louigi; Broutin, Nicolas; Devroye, Luc; Lugosi, Gábor. On combinatorial testing problems. Ann. Statist., Tome 38 (2010) no. 1, pp.  3063-3092. http://gdmltest.u-ga.fr/item/1283175989/