fwer_critical: Calculate the critical value and the marginal type-I error...

In PlatformDesign: Optimal Two-Period Multiarm Platform Design with New Experimental Arms Added During the Trial

Calculate the critical value and the marginal type-I error rate

Description

Calculate the critical value and the marginal type-I error rate given the number of experimental arms, the family-wise type I error rate and the correlation matrix of the z-statistics.

Usage

fwer_critical(ntrt, fwer, corMat, seed = 123)

Arguments

ntrt	the number of experimental arms in the trial
fwer	the family-wise error rate (FWER) to be controlled, default to be the same throughout the trial
corMat	the correlation matrix of the Z-test statistics
seed	an integer used in random number generation for numerically evaluating integration, default = 123

Details

Use the number of experimental arms, the family-wise type I error rate and the correlation matrix of the Z-test statistics to calculate the marginal type I error rate and the critical value.

Value

pairwise_alpha the marginal type-I error rate for the comparison between any of the experimental arm and its corresponding control

critical_val, the critical value for the comparison between any of the experimental arm and the corresponding controls

Other values returned are inputs.

Author(s)

Xiaomeng Yuan, Haitao Pan

References

Dunnett, C. W. (1955). A multiple comparison procedure for comparing several treatments with a control. Journal of the American Statistical Association, 50(272), 1096-1121.

Examples

corMat1 <- cor.mat(K=2, M = 2, n=107, n0=198, n0t = 43)$cormat
fwer_critical(ntrt=4, fwer=0.025, corMat=corMat1)
#
#$ntrt
#[1] 4
#
#$fwer
#[1] 0.025
#
#$corMat
#      [,1]      [,2]      [,3]      [,4]
#[1,] 1.0000000 0.3508197 0.2746316 0.2746316
#[2,] 0.3508197 1.0000000 0.2746316 0.2746316
#[3,] 0.2746316 0.2746316 1.0000000 0.3508197
#[4,] 0.2746316 0.2746316 0.3508197 1.0000000
#
#$pairwise_alpha
#[1] 0.006657461
#
#$critical_val
#[1] 2.475233

PlatformDesign documentation built on Dec. 28, 2022, 1:53 a.m.