coxph_mi: Multiple Imputation with Rubin's rule and Wild Bootstrap

In elong0527/smim: Survival Sensitivity Analysis using Multiple Imputation and Martingale

Description

Multiple Imputation with Rubin's rule and Wild Bootstrap

Usage

coxph_mi(
  fit,
  pattern,
  delta,
  n_mi,
  n_wb,
  cut_point,
  seed = NULL,
  validate = FALSE
)

Arguments

fit	a Cox model object, i.e., the result of coxph.
pattern	A integer vector of pattern indicator for each subject * 1 = observed event time * 2 = censoring at random * 3+ = censoring not at random
delta	A numeric vector of adjusted delta in the same length of pattern. * delta = 1 is for observed event time and censoring at random. * delta > 1 is for censoring time with hazard is larger than the hazard in CAR situation. * delta < 1 is for censoring time with hazard is smaller than the hazard in CAR situation.
n_mi	Number of multiple imputation
n_wb	Number of bootstrap replications
cut_point	Cutpoint of RMST
seed	Randomization seed
validate	Validation model, default is false. Only used for testing purpose.

elong0527/smim documentation built on May 5, 2021, 1:03 p.m.