A New Method for Testing Separate Families of Hypotheses (Q3707100)
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 3936190
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | A New Method for Testing Separate Families of Hypotheses |
scientific article; zbMATH DE number 3936190 |
Statements
1985
0 references
spline
0 references
significance level
0 references
testing separate families of hypotheses
0 references
parametric bootstrap
0 references
shrinking confidence regions
0 references
finite-sample behaviour
0 references
A New Method for Testing Separate Families of Hypotheses (English)
0 references
0.7691808938980103
0 references
0.7691808938980103
0 references
0.7597032189369202
0 references
0.7464584708213806
0 references
The well-known tests given by \textit{D. R. Cox} [Proc. 4th Berkeley Sympos. math. statist. Probab. 1, 105-123 (1961)] and \textit{D. A. Williams} [Biometrics 28, 23-32 (1970)] for the problem of testing separate families of hypotheses are not guaranteed to be asymptotically level \(\alpha\), even for very smooth problems. A new method for constructing a test for separate families is proposed. The parametric bootstrap idea is repeatedly applied over shrinking confidence regions. Properties of the method are derived and its finite- sample behaviour is studied in a number of examples.
0 references