Expanded confidence intervals (Q3761471)
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 4011653
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Expanded confidence intervals |
scientific article; zbMATH DE number 4011653 |
Statements
Expanded confidence intervals (English)
0 references
1985
0 references
threshold
0 references
one-tailed
0 references
studentized range
0 references
comparisons with control
0 references
two- sided Dunnett procedure
0 references
crucial points
0 references
expanded confidence interval
0 references
normal distributions
0 references
multiple comparisons
0 references
0.7998151779174805
0 references
0.7991833090782166
0 references
0.7825303673744202
0 references
0.7680071592330933
0 references
0.7671226263046265
0 references
A random variable \(X\) with density \(f(x-\theta)\) and p.d.f. F, \(\theta\in ({\underline \theta},{\bar \theta}]\), is considered. The ''crucial points'' are given by \({\underline \theta}=A_ 0<A_ 1<A_ 2<...<A_{k+1}={\bar \theta}.\) Let \(x\) be an observed value of \(X\), \((x-d,x+d]\) a confidence interval for \(\theta\). The expanded confidence interval \(I=((x-d)_ L,(x+d)_ U]\) is defined by: \((x-d)_ L=A_ i\) for \(A_ 0<A_ i<x-d\leq A_{i+1}\), and \(=x-d\) for \(x-d\leq A_ 1\); \((x+d)_ U\) is defined analogously. The author studies the properties of confidence intervals (''of confidence coefficients'') for various \(\delta\), where \(\delta =\min_{1\leq i<k}(A_{i+1}-A_ i)\). The infimum of the confidence coefficient of the I is given by \(F(d)-F(-d-\delta)\), if \(f(x)\) is symmetric about zero and unimodal (at zero). Special attention is devoted to hypothesis testing using expanded confidence intervals, to normal distributions and to multiple comparisons.
0 references