FinalCI: calculate confidence intervals at the end of the study

View source: R/FinalCI.R

FinalCIR Documentation

calculate confidence intervals at the end of the study

Description

calculate confidence intervals at the end of the study

Usage

FinalCI(
  Info.d,
  Info.i,
  ck,
  ck.unrestricted,
  lk,
  uk,
  reason.interim,
  kMax,
  conf.level,
  estimate,
  optimizer = "optimise",
  method,
  bindingFutility,
  cNotBelowFixedc,
  continuity.correction,
  tolerance,
  FCT.p_value
)

Arguments

Info.d

Information at all decision analyses up to stage where study was stopped

Info.i

Information at all interim analyses up to stage where study was stopped

ck, ck.unrestricted

decision boundaries for all decision analyses up to stage where study was stopped (should include final boundary if stopped at final analysis). ck is possibly with restriction (when cNotBelowFixedc=TRUE) and ck.unrestricted always without.

lk

lower bounds up to stage where study was stopped

uk

upper bounds up to stage where study was stopped

reason.interim

motivation for stopping or continuing at interim. Use to handle special cases (skipped interim because reach Imax, ...)

kMax

maximum number of analyses

conf.level

confidence level (to get a 100*(1-alpha)% CI)

estimate

naive estimate (e.g. using ML or REML).

optimizer

the observed treatment estimate at decision

method

method 1, 2 or 3

bindingFutility

[logical] whether the futility stopping rule is binding.

cNotBelowFixedc

[logical] whether the value c at the decision analysis can be below that of a fixed sample test (H & J page 10)

continuity.correction

[logical] whether to add the probability of stopping between ck and ck.uncorrected to ensure continuity of the p-value across stages. When used the p-value will always be greater than this probability of stopping bettwen ck and ck.uncorrected.

tolerance

[numeric] acceptable discrepancy to the objective level when evaluating the confidence intervals and median unbiased estimate.

FCT.p_value

[function] function used to compute the p-value.


paulowhite/DelayedGSD documentation built on Nov. 1, 2023, 5:36 p.m.