Based on a pre-planned graphical multiple comparison procedure, construct a valid multiple level alpha test that conserves the family wise error in the strong sense regardless of any trial adaptations during an unblinded interim analysis. - Implementation of adaptive procedures is still in an early stage and may change in the near future
secondStageTest(
interim,
select,
matchCE = TRUE,
zWeights = "reject",
G2 = interim@preplanned
)A function of signature function(z2) with arguments
z2 a numeric vector with second stage z-scores (Z-scores of
dropped hypotheses should be set no NA)
that returns objects of class gMCPResult.
An object of class gPADInterim.
A logical vector giving specifying which hypotheses are carried forward to the second stage
Logical specifying whether second stage weights should be computed proportional to corresponding PCEs
Either "reject","accept", or "strict" giving the rule what should be done in cases where none of the selected hypotheses has positive second stage weight.
An object of class graphMCP laying down the rule to
compute second stage weights. Defaults to pre-planned graph.
Florian Klinglmueller float@lefant.net
For details see the given 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. https://www.meduniwien.ac.at/fwf_adaptive/papers/bretz_2009_22.pdf
Bretz F., Posch M., Glimm E., Klinglmueller F., Maurer W., Rohmeyer K. (2011): Graphical approaches for multiple endpoint problems using weighted Bonferroni, Simes or parametric tests - to appear.
Posch M, Futschik A (2008): A Uniform Improvement of Bonferroni-Type Tests by Sequential Tests JASA 103/481, 299-308
Posch M, Maurer W, Bretz F (2010): Type I error rate control in adaptive designs for confirmatory clinical trials with treatment selection at interim Pharm Stat 10/2, 96-104
graphMCP, doInterim
## Simple successive graph (Maurer et al. 2011)
## two treatments two hierarchically ordered endpoints
a <- .025
G <- simpleSuccessiveI()
## some z-scores:
p1=c(.1,.12,.21,.16)
z1 <- qnorm(1-p1)
p2=c(.04,1,.14,1)
z2 <- qnorm(1-p2)
v <- c(1/2,1/3,1/2,1/3)
intA <- doInterim(G,z1,v)
## select only the first treatment
fTest <- secondStageTest(intA,c(1,0,1,0))
Run the code above in your browser using DataLab