asd (version 2.2)

hyp.test: Closed Testing for ASD

Description

Implements the closure principle (Marcus et al., 1976) for controlling the familywise type I error rate in ASD.

Usage

hyp.test(comb.test, level = level, full.hyp = FALSE)

Arguments

comb.test

Output from function combn.test

level

Test level (default=0.025)

full.hyp

Logical indicating whether the full set of intersection hypotheses should be reported; default FALSE

Value

reject

Matrix indicating whether elementary hypotheses have been rejected

all.rejects

Matrix indicating rejections for each intersection hypothesis, if full.hyp=TRUE

all.hyp

Matrix labelling each intersection hypothesis, if full.hyp=TRUE

Details

In order to control the familywise type I error rate in the strong sense at the pre-specified level $\alpha$ the closure principle (Marcus et al., 1976) is applied. This means that an individual null hypothesis is rejected if and only if all intersection hypotheses are also rejected at level $\alpha$.

References

Marcus R, Peritz E, Gabriel KR. On closed testing procedures with special reference to ordered analysis of variance. Biometrika 1976;63:655-660.

See Also

treatsel.sim,dunnett.test, combn.test, select.rule, simeans.binormal

Examples

stage1 <- dunnett.test(c(0.75,1.5,2.25))
stage2 <- dunnett.test(c(0.15,1.75,2.15))
comb.test <- combn.test(stage1,stage2,weight=0.5)
hyp.test(comb.test,level=0.025,full.hyp=FALSE)

# more output
hyp.test(comb.test,level=0.025,full.hyp=TRUE)