Fuzzy and Randomized Confidence Intervals and P-Values
Geyer, Charles J. ; Meeden, Glen D.
Statist. Sci., Tome 20 (2005) no. 1, p. 358-366 / Harvested from Project Euclid
The optimal hypothesis tests for the binomial distribution and some other discrete distributions are uniformly most powerful (UMP) one-tailed and UMP unbiased (UMPU) two-tailed randomized tests. Conventional confidence intervals are not dual to randomized tests and perform badly on discrete data at small and moderate sample sizes. We introduce a new confidence interval notion, called fuzzy confidence intervals, that is dual to and inherits the exactness and optimality of UMP and UMPU tests. We also introduce a new P-value notion, called fuzzy P-values or abstract randomized P-values, that also inherits the same exactness and optimality.
Publié le : 2005-11-14
Classification:  Confidence interval,  P-value,  hypothesis test,  uniformly most powerful unbiased (UMP and UMPU),  fuzzy set theory,  randomized test
@article{1137076652,
     author = {Geyer, Charles J. and Meeden, Glen D.},
     title = {Fuzzy and Randomized Confidence Intervals and P-Values},
     journal = {Statist. Sci.},
     volume = {20},
     number = {1},
     year = {2005},
     pages = { 358-366},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1137076652}
}
Geyer, Charles J.; Meeden, Glen D. Fuzzy and Randomized Confidence Intervals and P-Values. Statist. Sci., Tome 20 (2005) no. 1, pp.  358-366. http://gdmltest.u-ga.fr/item/1137076652/