onestep_txshift {txshift}

One-Step Estimate of Counterfactual Mean of Stochastic Shift Intervention

Description

One-Step Estimate of Counterfactual Mean of Stochastic Shift Intervention

Usage

onestep_txshift(
  data_internal,
  C_samp = rep(1, nrow(data_internal)),
  V = NULL,
  delta,
  samp_estim,
  gn_cens_weights,
  Qn_estim,
  Hn_estim,
  eif_reg_type = c("hal", "glm"),
  samp_fit_args,
  ipcw_efficiency = TRUE
)

Arguments

data_internal

A data.table constructed internally by a call to txshift. This contains the data elements needed for computing the one-step estimator.

C_samp

A numeric indicator for whether a given observation was included in the second-stage sample, used to compute an IPC-weighted one-step estimator in cases where two-stage sampling is performed. Default assumes no censoring due to sampling.

V

The covariates that are used in determining the sampling procedure that gives rise to censoring. The default is NULL and corresponds to scenarios in which there is no censoring (in which case all values in the preceding argument C must be uniquely 1. To specify this, pass in a NAMED list identifying variables amongst W, A, Y that are thought to have played a role in defining the sampling/censoring mechanism (C).

delta

A numeric value indicating the shift in the treatment to be used in defining the target parameter. This is defined with respect to the scale of the treatment (A).

samp_estim

An object providing the value of the censoring mechanism evaluated across the full data. This object is passed in after being constructed by a call to the internal function est_samp.

gn_cens_weights

TODO: document

Qn_estim

An object providing the value of the outcome evaluated after imposing a shift in the treatment. This object is passed in after being constructed by a call to the internal function est_Q.

Hn_estim

An object providing values of the auxiliary ("clever") covariate, constructed from the treatment mechanism and required for targeted minimum loss estimation. This object object should be passed in after being constructed by a call to the internal function est_Hn.

eif_reg_type

Whether a flexible nonparametric function ought to be used in the dimension-reduced nuisance regression of the targeting step for the censored data case. By default, the method used is a nonparametric regression based on the Highly Adaptive Lasso (from hal9001). Set this to "glm" to instead use a simple linear regression model. In this step, the efficient influence function (EIF) is regressed against covariates contributing to the censoring mechanism (i.e., EIF ~ V | C = 1).

samp_fit_args

A list of arguments, all but one of which are passed to est_samp. For details, consult the documentation for est_samp. The first element (i.e., fit_type) is used to determine how this regression is fit: "glm" for generalized linear model, "sl" for a Super Learner, and "external" for a user-specified input of the form produced by est_samp.

ipcw_efficiency

Whether to invoke an augmentation of the IPCW-TMLE procedure that performs an iterative process to ensure efficiency of the resulting estimate. The default is TRUE; set to FALSE to use an IPC-weighted loss rather than the IPC-augmented influence function.

Details

Invokes the procedure to construct a one-step estimate of the counterfactual mean under a modified treatment policy.

Value

S3 object of class txshift containing the results of the procedure to compute a one-step estimate of the treatment shift parameter.