Test-wise alpha necessary to control the overall type I error at a specified level (0.05 by default)
getAlphaPerTest(
nevents,
totalAlpha = 0.05,
pH0 = 0.5,
alpha.interval = c(10^(-10), 0.05)
)
vector with number of events at which an interim analysis is done
Overall type I error, 0.05 by default
proportion of events in the experimental arm under the null hypothesis, typically based on randomization ratio (e.g. 0.5 for a 1:1 randomization)
Range for test-wise alpha, c(10^(-10),0.05) by default
Test-wide alpha
getAlphaPerTest(nevents = c(10,50,100), totalAlpha = 0.05, pH0 = 0.5)
#> [1] 0.03245429