jt_design: Compute the design by Jennison and Turnbull (2015)

In MatheMax/OptReSample: Computing Optimal Adaptive Designs

Description

jt_design computes the design proposed by Jennison and Turnbull (2015). It is based on the inverse normal combination test. The n_2-function is termined in order to minimize expected sample size under the alternative.

Usage

jt_design(theta, sigma, n1 = NA, lb_n2, alpha, beta)

Arguments

theta	Effect size for which the expected sample size should be minimized
sigma	Known standard deviation
n1	First stage sample size. Write NA to obtain the optimal n_1.
lb_n2	Lower bound for the stage two sample size due to e.g. delayed response
alpha	Maximal type I error rate
beta	Maximal type II error rate

Details

Note that the sample sizes must be given per group. The total sample size is 2*(n_1 + n_2).

MatheMax/OptReSample documentation built on May 5, 2019, 8:14 a.m.