rr.msm.jn.ic {amp}

Estimate both the parameter, and the influence curves used for estimating the projected risk ratio from a working marginal structural mean model. This function only uses elastic net for the estimation (rather than also using other learners).

Description

The function takes a dataset, and will returns the estimate for the parameter, and or the estimated influence curve at each observation. The first column of obs_data should be the binary outcome of interest.

Usage

rr.msm.jn.ic(obs_data, what = "both", control = NULL)

Arguments

obs_data

the observed data. The first column should be the outcome.

what

the desired return value. Should be one of '"ic"' (influence curve), '"est"' (estimate), or '"both"'.

control

any other control parameters to be passed to the estimator.

Details

Most of the code for this function has been copied and slightly modified from the ltmle package and is copyrighted by Joshua Schwab under the terms of the GPL-2 license.

Obtain an estimator of the probability delta = 1 given w

Value

If 'what' is

- '"est"', then return the estimated parameter.

- '"ic"', then return the estimated IC of the parameter estimate.

- '"both"', then return both the parameter estimate and corresponding estimated IC.

Examples

#not run (make sure to load in SuperLearner if running)
# set.seed(1010)
# fake_dat <- data.frame(y = rbinom(100, size = 1, prob = 0.5),
#                        delta = rbinom(100, size =  1, prob = 0.5),
#                        w = matrix(rnorm(500), ncol = 5))
# rr.msm.jn.ic(fake_dat)