permutation_CER2: Permutation Conditional Error Rate
In floatofmath/adaperm: Permutation tests for adaptive designs

Description Usage Arguments Details Value Author(s)

Computes the conditional type I error rate of a pre-planned permutation test in a two-stage adaptive design. We condition on the observed first stage data and treatment assignment as well as the observed second stage data as well as the observations from the extension of the trial.

1
2
3

permutation_CER2(x1, g1, x2, x3, stat = sumdiff, permutations = 1000,
  subsamples = 1000, alpha = 0.025, g2 = rep(c(-1, 1), each =
  length(x2)/2), restricted = TRUE, ...)

`x1`	vector of preplanned first stage observations
`g1`	vector of first stage treatment assignments
`x2`	vector of preplanned second stage observations
`x3`	vector of observations from the extended trial
`stat`	function computing the test statistic (see Details)
`permutations`	number of permutations (rerandomizations) used to compute unconditional and conditional permutation distributions
`subsamples`	number of (second stage) subsamples used for conditional error rate estimation
`alpha`	pre-fixed significance level
`g2`	template vector for second stage treatment assignments
`restricted`	should group sizes be treated fixed
`...`	additional options to `stat`

Based on the first stage data and treatment assignments one may perform sample size reassassment - and possibly other trial modifications - as long as the (preplanned) second stage sample size is not reduced.

stat needs to be a function of the form function(x,g,...) returning a numeric of length one. Possible options are sumdiff, meandiff, zstat

For the moment, we assume that observations are randomized using random allocation blocked by stages, (i.e. we resample using sample (g1)). g2 does not have to be the actual second stage treatment assignments but just one possible example randomization, that fixes the treatment group sizes.

numeric value of the conditional error rate

Florian Klinglmueller

floatofmath/adaperm documentation built on May 16, 2019, 1:18 p.m.