A family of variable stage size multistage tests of simple hypotheses is described, based on efficient multistage sampling procedures. Using a loss function that is a linear combination of sampling costs and error probabilities, these tests are shown to minimize the integrated risk to second order as the costs per stage and per observation approach zero. A numerical study shows significant improvement over group sequential tests in a binomial testing problem.
Publié le : 2007-10-14
Classification:
Multistage,
hypothesis testing,
asymptotic optimality,
group sequential,
62L10,
91A20
@article{1194461723,
author = {Bartroff, Jay},
title = {Asymptotically optimal multistage tests of simple hypotheses},
journal = {Ann. Statist.},
volume = {35},
number = {1},
year = {2007},
pages = { 2075-2105},
language = {en},
url = {http://dml.mathdoc.fr/item/1194461723}
}
Bartroff, Jay. Asymptotically optimal multistage tests of simple hypotheses. Ann. Statist., Tome 35 (2007) no. 1, pp. 2075-2105. http://gdmltest.u-ga.fr/item/1194461723/