calc_expEvents: Calculates the expected number of events
In Knusprikus/BSSRed: Blinded Sample Size Reestimation For Event Driven Trials

Description Usage Arguments Details Value Examples

!!!!!!!!!!!!! shall not be exported !!!!!!!!!!!!!!!!!!! Calculates the expected number of events of a population given their respective recruiting timepoint, assumed survival time of the first group, the hazard ratio and ce censor rate. If the distributions are ....

calc_expEvents(
  N,
  theta,
  L = Inf,
  tn,
  lambda,
  sigma = 1,
  distS = "exponential",
  gamma,
  kappa = 1,
  distC = "exponential",
  col_Sums = TRUE
)

`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`).
`theta`	a number greater 0 defining the assumed hazard ratio.
`L`	a number greater 0 defining the timepoint for administrative censoring. If no administrative censoring is planned `L=Inf`. Default is L=Inf.
`tn`	a numeric vector defining the recruitment timepoint. Default is `0:(NROW(N)-1)`.
`lambda`	a number greater 0 defining the hazard rate for the survival process of group1. For the parametrization see `details`.
`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 hazard rate for the censor process.
`col_Sums`	a boolen value. If set to FALSE, the function will return the expected number of events per recruitment batch at time T instead of the sums per group.

With constant hazard rate λ and shape parameter σ the survival function for the weibull event process is given by

S ( t ) = exp{ λ t ^ σ },

with density function

f ( t ) = λ σ t ^ { σ - 1 } exp{ λ t ^ σ }.

a vector containing the expected number of events per group at time L.

# with a recruitment time of 5 month and recruiting 30 Patients per month
# assuming a hazard ratio of .7 and a rate of two year survival of .7
# as well as a two year censor rate of .2 under exponential distribution
# and administrative censoring after 10 month.
N <- matrix(rep(50,10),ncol=2)
calc_expEvents(N=N ,theta = .7, L=10, lambda=-log(.7)/24, gamma=-log(.8)/24)