weighted.tmle.separate: IPCWTMLE2Phase Main IPCW-TMLE function for two-phase studies

In brendalewisprice/IPCWTMLE2Phase: Inverse Probability of Censoring Weighted Targeted Minimum Loss-based Estimation for Two-Phase STudies

Description

IPCW-TMLE function to estimate E[E[Y|A=1]] and E[E[Y|A=0]] separately, and to also estimate the marginalized risk difference, relative risk, and odds ratio. This function assumes that the full phase one data set is input with weights equal to 0 for those observations not selected for the second phase. An indicator variable, delta, should equal 1 for all phase-two data and 0 for all other observations (non-phase-two).

Usage

weighted.tmle.separate(w, a, y, delta, Q.SL.library, g.SL.library, wgts,
  max.wgt = Inf)

Arguments

w	The covariates
a	The treatment or exposure binary variable.
y	The outcome
delta	Indicator for inclusion into the second phase
Q.SL.library	SuperLearner library of algorithms for Q estimation
g.SL.library	SuperLearner library of algorithms for g estimation
wgts	Observation level weights; should be 0 for delta=0

Value

Estimates of marginalized risk difference, RR, and OR

Examples

weighted.tmle.separate

brendalewisprice/IPCWTMLE2Phase documentation built on Nov. 4, 2019, 8:12 a.m.