Usage
washb_tmle(Y, tr, strat, W, id, contrast, prtr = c(0.5, 0.5),
modelfit = "glm", family = "gaussian", SL.library = c("SL.mean",
"SL.glm", "SL.bayesglm", "SL.gam", "SL.glmnet"), seed = NULL)
Arguments
- Y
- Outcome variable (continuous, such as LAZ, or binary, such as diarrhea)
- tr
- Binary treatment group variable, comparison group first
- strat
- Stratification variable (here: block)
- W
- Data frame that includes adjustment covariates
- id
- ID variable for independent units (e.g., cluster)
- contrast
- Vector of length 2 that includes the tr groups to contrast
- prtr
- Vector of length 2 that includes the probability of receiving each treatment in the contrast argument for the main trial, the control is double sized (allocation 2:1) so control vs. active intervention should be prtr=c(0.67,0.33). in contrasts where the arms are equally sized (allocation 1:1), then prtr=c(0.5,0.5).
- modelfit
- String variable of value "glm" for a glm prediction of E(Y|A,W) and E(A|W), or "sl" for a SuperLearner prediction.
- family
- Outcome family: gaussian (continuous outcomes, like LAZ) or binomial (binary outcomes like diarrhea or stunting)
- seed
- A seed for the pseudo-random CV split for perfectly reproducible results
- SL.Library
- Library of algorithms to include in the SuperLearner (pre-specified defaults are encoded above)
Description
NOTE: washb_tmle has not been fully implemented in this package version. Do not use yet.
Details
# Targeted maximum likelihood estimation (TMLE) for the WASH Benefits intention to treat (ITT) analyses
The function does the following:
Examples