It is well known that anomalies are sometimes observed when using the likelihood ratio test (LRT) for testing restricted hypotheses in a normal model. This paper considers a general framework for these anomalies to occur. We provide a condition, that relates the null and the alternative hypotheses, under which the dominance of the LRT is obtained. Conditions are also given which guarantee the equivalence between the LRT and a simpler test. The situations of known and unknown variances are considered and examples are given to illustrate the results.
Publié le : 1992-12-14
Classification:
Restricted inference,
likelihood ratio test,
obliqueness,
62F03,
62H15
@article{1176348904,
author = {Menendez, J. A. and Rueda, C. and Salvador, B.},
title = {Dominance of Likelihood Ratio Tests under Cone Constraints},
journal = {Ann. Statist.},
volume = {20},
number = {1},
year = {1992},
pages = { 2087-2099},
language = {en},
url = {http://dml.mathdoc.fr/item/1176348904}
}
Menendez, J. A.; Rueda, C.; Salvador, B. Dominance of Likelihood Ratio Tests under Cone Constraints. Ann. Statist., Tome 20 (1992) no. 1, pp. 2087-2099. http://gdmltest.u-ga.fr/item/1176348904/