safestats (version 0.8.7)

selectivelyContinueTTestCombineData: Selectively Continue Experiments that Did Not Lead to a Null Rejection for a (Safe) T-Test

Description

Helper function used in the vignette.

Usage

selectivelyContinueTTestCombineData(
  oldValues,
  valuesType = c("eValues", "pValues"),
  designObj = NULL,
  alternative = c("twoSided", "greater", "less"),
  oldData,
  deltaTrue,
  alpha = NULL,
  n1Extra = NULL,
  n2Extra = NULL,
  seed = NULL,
  paired = FALSE,
  muGlobal = 0,
  sigmaTrue = 1,
  moreMainText = ""
)

Value

a list that includes the continued s or p-values based on the combined data, and a list of the combined data.

Arguments

oldValues

vector of e-values or p-values.

valuesType

character string either "eValues" or "pValues".

designObj

a safeDesign object obtained from designSafeT, or NULL if valuesType equal "pValues".

alternative

a character string specifying the alternative hypothesis must be one of "twoSided" (default), "greater" or "less".

oldData

a list of matrices with names "dataGroup1" and "dataGroup2".

deltaTrue

numeric, the value of the true standardised effect size (test-relevant parameter).

alpha

numeric in (0, 1) that specifies the tolerable type I error control --independent of n-- that the designed test has to adhere to. Note that it also defines the rejection rule e10 > 1/alpha.

n1Extra

integer, that defines the additional number of samples of the first group. If NULL and valuesType equals "eValues", then n1Extra equals designObj$nPlan[1].

n2Extra

optional integer, that defines the additional number of samples of the second group. If NULL, and valuesType equals "eValues", then n2Extra equals designObj$nPlan[2].

seed

To set the seed for the simulated data.

paired

logical, if TRUE then paired t-test.

muGlobal

numeric, the true global mean of a paired or two-sample t-test. Its value should not matter for the test. This parameter is treated as a nuisance.

sigmaTrue

numeric > 0,the true standard deviation of the data. Its value should not matter for the test.This parameter treated is treated as a nuisance.

moreMainText

character, additional remarks in the title of the histogram.

Examples

alpha <- 0.05
mIter <- 1000L

designObj <- designSafeT(deltaMin=1, alpha=alpha, beta=0.2, nSim=100)
oldData <- generateNormalData(nPlan=designObj$nPlan, deltaTrue=0, nSim=mIter, seed=1)

eValues <- vector("numeric", length=mIter)

for (i in seq_along(eValues)) {
  eValues[i] <- safeTTest(x=oldData$dataGroup1[i, ], designObj=designObj)$eValue
}

# First run: 8 false null rejections
sum(eValues > 1/alpha)

continuedSafe <- selectivelyContinueTTestCombineData(
  oldValues=eValues, designObj=designObj, oldData=oldData,
  deltaTrue=0, seed=2)

# Second run: 1 false null rejections
sum(continuedSafe$newValues > 1/alpha)

# Third run: 0 false null rejections
eValues <- continuedSafe$newValues
oldData <- continuedSafe$combinedData
continuedSafe <- selectivelyContinueTTestCombineData(
  oldValues=eValues, designObj=designObj, oldData=oldData,
  deltaTrue=0, seed=3)
sum(continuedSafe$newValues > 1/alpha)