onestep_single_t: One-step TMLE estimator for survival at specific time point

Description Usage Arguments Examples

View source: R/1_surv_onestep.R

Description

One-step TMLE estimator for survival at specific time point

Usage

1
2
3
4
5
onestep_single_t(dat, tk, dW = rep(1, nrow(dat)), SL.trt = c("SL.glm",
  "SL.step", "SL.earth"), SL.ctime = c("SL.glm", "SL.step", "SL.earth"),
  SL.ftime = c("SL.glm", "SL.step", "SL.earth"), maxIter = 300,
  epsilon_step = 0.001, tol = 1/nrow(dat), T.cutoff = NULL,
  verbose = FALSE)

Arguments

dat

data.frame with columns T, A, C, W. All columns with character "W" will be treated as baseline covariates.

tk

time point to compute survival probability

dW

binary input vector specifying dynamic treatment (as a function output of W)

SL.trt

SuperLearner library for fitting treatment regression

SL.ctime

SuperLearner library for fitting censoring regression

SL.ftime

SuperLearner library for fitting conditional hazard regression

maxIter

maximal number of recursion for one-step

epsilon_step

step size for one-step recursion

tol

tolerance for optimization

T.cutoff

manual right censor the data; remove parts dont want to esimate

verbose

to print log-likelihood value during optimzation

Examples

1
# TO DO

wilsoncai1992/onestep.survival documentation built on Dec. 17, 2017, 12:07 p.m.