btEstimate: Blinded Study Time Estimation
In Knusprikus/BSSRed: Blinded Sample Size Reestimation For Event Driven Trials

Description Usage Arguments Value Examples

!!!!!!!!!!!!! shall not be exported !!!!!!!!!!!!!!!!!!!

btEstimate(
  dfSurv,
  theta.star,
  N,
  tn,
  M,
  tm,
  nEvents,
  L = L,
  sigma = sigma,
  distS = distS,
  gamma = gamma,
  kappa = kappa,
  distC = distC
)

`dfSurv`	a data.frame containing the variables: status: (0,1) := (no event,event) time: time to status group: grouping variable for a
`theta.star`	a number greater 0 defining the assumed hazard ratio.
`N`	a two column matrix containing the size of both recruitment groups per row, representing the number of recruited patients per group and timepoint (`tn`).
`tn`	a vecotr of length = `NROW(N)` defining the timepoint of recruiting.
`M`	a two column matrix containing the size of both recruitment groups per row, representing the number of future recruited patients per group and timepoint (`tm`). If not defined it will be taken from the last row of `N`.
`tm`	a vecotr of length = `NROW(M)` defining the timepoint of future recruiting. If not defined it will be interpolated from the last two entries of `tn`.
`nEvents`	a number greater 0 defining the targeted number of events to achieve the planned power.
`L`	a number defining the timepoint for administrative censoring. If no administrative censoring is planned `L=Inf`. Default is L=Inf.
`sigma, kappa`	a number greater 0 defining the shape parameter for the survival- and the censor process. Only needed if distS or distC is set to weibull. Default is 1 resulting in an exponential distribution.
`distS, distC`	a character string defining the distribution of the survival- and the censor process. Default is 'exponential'.
`gamma`	a number greater 0 defining the overall rate for the censor process.
`from`	a number defining the minimum recruitment time.
`lambda`	a number greater 0 defining the rate for the survival process of group1. For the parametrization see `details`.

a number defining the minimum recruitmenttime, given the recruitments N, rn and the blinded estimation of the event rates from the supplied data.frame.

N <- matrix(rep(50,20),ncol=2)
tn <- 0:(NROW(N)-1)
dfSurv <- simSurvData(N=N, lambda=exp(.7)/12, sigma=1, theta=.7, gamma=exp(.8)/12, kappa=1, distS = "exponential",distC = "exponential",L=10)
btEstimate(dfSurv=dfSurv, theta.star=.7, N=N, tn=tn, nEvents=40, L=20,
           sigma=1, distS="exponential",
           gamma=-log(.3)/24, kappa=1, distC="exponential")