pickwin: Pick-the-winner design

In raredd/rawr: rawr miscellaneous

Pick-the-winner design

Description

Calculate sample size of Simon's pick winner design or determine the probability of alternative hypothesis to obtain target power.

Usage

pickwin(
  n = NULL,
  k = 2L,
  p0 = NULL,
  p1 = NULL,
  power = NULL,
  tol = .Machine$double.eps^0.25
)

Arguments

n	sample size per arm
k	number of arms
p0, p1	null and alternative response rates, respectively
power	probability of choosing the winning arm
tol	the desired accuracy (convergence tolerance); see uniroot

Details

These calculations use method proposed by R. Simon for pick-the-winner design for randomized phase II clinical trials.

Exactly one of the parameters n, power, and p1 must be NULL.

This design is not a hypothesis test, nor is there a formal comparison between arms. Therefore, the type I error rate does not apply.

Value

An object of class power.htest with the following elements:

n	sample size per arm
arm	number of arms
n.total	total sample size for the study

References

Adapted from power.ctepd::n.pick.winner, Bingshu E. Chen

Simon, R., R.E. Wittes, S.S. Ellenberg. Randomized phase II clinical trials. Cancer Treat Rep. 69:1375-1381, 1985.

See Also

power.ctepd::n.pick.winner; pselect; pickwin;

Examples

pickwin(20, 2, 0.25, 0.5)
pickwin(NULL, 2, 0.25, 0.5, 0.95)

pickwin(20, 3, 0.25, 0.5)$power
pickwin(20, 4, 0.25, 0.5)$power

raredd/rawr documentation built on Feb. 25, 2025, 1:48 p.m.