inference: Exact limits from outcome

In CLAST: Exact Confidence Limits after a Sequential Trial

Exact limits from outcome

Description

Calculates exact upper and lower confidence limits from the outcome of a multi-stage group sequential trial.

Usage

inference(n, a, b, y, alpha, type)

Arguments

n	Design vector of planned sample sizes
a	Design vector of lower futility boundaries
b	Design vector of upper superiority boundaries
y	Actual outcome of experiment
alpha	exact coverage error
type	character, either "LR", "CP", "JT" or "ML"

Value

list with components

lower	exact lower limit, type dependent
upper	exact lower limit, type dependent
est	ML estimate
type	selected ordering function "LR", "CP", "JT" or "ML"

Author(s)

Chris J. Lloyd

References

Lloyd, C.J. (2020) Exact confidence limits after a group sequential single arm binary trial. Statistics in Medicine, Volume 38, 2389-2399.

Examples

# Example from table 1 of Lloyd (2020)
  n=c(5,6,5,9)
  a=c(2,4,5,12)
  b=c(5,9,11,13)
  y=c(4,2,5) # This is the actual outcome. The trial stopped for
#              superiority at stage m=3 with s=11 successes.
  inference(n,a,b,y,type="CP")
# Results happen to be identical for type="LR" but different
# for the alternative ordering functions "JT" and "ML"

CLAST documentation built on April 8, 2022, 9:06 a.m.