No label defined (Q3713341)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3942738
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 3942738 |
Statements
1985
0 references
empirical processes
0 references
simple null hypothesis
0 references
sequence of contiguous alternatives
0 references
quantile processes
0 references
limiting asymptotic power
0 references
locally most powerful tests
0 references
counterexample
0 references
power equality
0 references
0.7691527605056763
0 references
[For the entire collection see Zbl 0552.00002.] The problem is considered of testing a simple null hypothesis about a multidimensional parameter versus a sequence of contiguous alternatives. It is shown that in two important examples, that of empirical and quantile processes, the limiting asymptotic power of locally most powerful tests is the same as the power of the best test based on the stochastic process which arises as a distributional limit from the contiguous alternatives. It is claimed that in this case the latter is easier to compute, although no general results are produced which enable one to know when the power equality holds. A counterexample is given which shows that the power equality is not universal. The counterexample involves independent and identically distributed random variables with mean and variance dependent on an unknown parameter.
0 references