simMBIA: Maximum bias of the estimate mean difference following...

In floatofmath/blindConfidence: Simulation of estimation after blinded interim analysis

Description

simMBIA computes, via Monte-Carlo simulation, the bias for estimates of the mean difference if the blinded sample size reassessment rule is chosen such that the bias is maximised.

Usage

simMBIA(delta = 0, n1 = 2, n2max = Inf, sigma = 1, runs = 100)

Arguments

delta	true mean difference
n1	first stage per group sample size
n2max	maximum second stage sample size
sigma	true standard deviation
runs	number of simulted trials

Value

A vector following components:

m.bias	Maximum positiv mean bias that can be attained for a given parameter setting
m.bias.n	Maximum negative mean bias that can be attained

floatofmath/blindConfidence documentation built on May 16, 2019, 1:20 p.m.