mfc.default: minimum failure count in RASMART subject to a lower bound of...

In wangjunyao0121/RASMART: Response Adaptive Sequencial Multiple Assignments Randomized Trial

Description

minimum failure count in RASMART subject to a lower bound of the power

Usage

## Default S3 method:
mfc(
  x = c(0.5, 0.5, 0.5),
  pi1 = c(0.4, 0.35, 0.25),
  pi2 = c(0.35, 0.32, 0.36, 0.56, 0.37, 0.4),
  P = c(1/3, 1/3, 1/3),
  n = 365,
  power = 0.8,
  alpha.step = 0.8,
  lambda_seq = seq(1, 100, by = 0.5),
  ...
)

Arguments

x	initial second stage randomization probability in SMART
pi1	the respond rate of treatment 1, 2, 3 in stage I in SMART
pi2	the respond rate for patients who do not respond to a treatment in stage I, but respond to another treatment in stage II
P	Randomization probability in stage I
n	Sample size in SMART
power	the minimum controlled power for RASMART
alpha.step	The step size to search a small alpha for satisfying the Lipschitz constant such that f(x) <= f(x_0) + df(x_0)(x-x_0) + 1/(2alpha)||x-x_0||^2. Default is 0.8.
lambda_seq	The penalty parameter to minimize the lagrangian function.
...	other arguments related to mfc

Examples

x = c(0.3, 0.3, 0.3)
pi1 = c(0.4, 0.35, 0.25)
pi2 = c(0.35, 0.32, 0.36, 0.56, 0.37, 0.4)
P = c(1/3, 1/3, 1/3)
n = 365
power = 0.8
alpha.step = 0.8
lambda_seq = seq(1,50, by = 1)

mfc(x, pi1, pi2, P, n, power, alpha.step, lambda_seq)

wangjunyao0121/RASMART documentation built on Dec. 23, 2021, 5:07 p.m.