simConfint {gMCP}

Simultaneous confidence intervals for sequentially rejective multiple test procedures

Description

Calculates simultaneous confidence intervals for sequentially rejective multiple test procedures.

Usage

	simConfint(object, pvalues, confint, alternative=c("less", "greater"), 
             estimates, df, alpha=0.05, mu=0)

Arguments

object

A graph of class graphMCP.

pvalues

A numeric vector specifying the p-values for the sequentially rejective MTP.

confint

One of the following: A character string "normal", "t" or a function that calculates the confidence intervals. If confint=="t" the parameter df must be specified. If confint is a function it must be of signature ("character", "numeric"), where the first parameter is the hypothesis name and the second the marginal confidence level (see examples).

alternative

A character string specifying the alternative hypothesis, must be "greater" or "less".

estimates

Point estimates for the parameters of interest.

df

Degree of freedom as numeric.

alpha

The overall alpha level as numeric scalar.

mu

The numerical parameter vector under null hypothesis.

Details

For details see the given references.

Value

A matrix with columns giving lower confidence limits, point estimates and upper confidence limits for each parameter. These will be labeled as "lower bound", "estimate" and "upper bound".

Author(s)

Kornelius Rohmeyer [e-mail]

References

Frank Bretz, Willi Maurer, Werner Brannath, Martin Posch: A graphical approach to sequentially rejective multiple test procedures. Statistics in Medicine 2009 vol. 28 issue 4 page 586-604. http://www.meduniwien.ac.at/fwf_adaptive/papers/bretz_2009_22.pdf

See Also

graphMCP

Examples

est <- c("H1"=0.860382, "H2"=0.9161474, "H3"=0.9732953)
# Sample standard deviations:
ssd <- c("H1"=0.8759528, "H2"=1.291310, "H3"=0.8570892)

pval <- c(0.01260, 0.05154, 0.02124)/2

simConfint(BonferroniHolm(3), pvalues=pval, 
		confint=function(node, alpha) {
			c(est[node]-qt(1-alpha,df=9)*ssd[node]/sqrt(10), Inf)
		}, estimates=est, alpha=0.025, mu=0, alternative="greater")

# Note that the sample standard deviations in the following call
# will be calculated from the pvalues and estimates.
ci <- simConfint(BonferroniHolm(3), pvalues=pval, 
		confint="t", df=9, estimates=est, alpha=0.025, alternative="greater")
ci
	
plotSimCI(ci)