tune_theta: Estimate the penalty parameter for Step 2 of Virtual Twins

Description

Permutes data under the null hypothesis of a constant treatment effect and calculates the MNPP on each permuted data set. The 1 - alpha quantile of the distribution is taken.

Usage

tune_theta(
  data,
  Trt,
  Y,
  zbar,
  step1,
  step2,
  threshold,
  alpha0,
  p_reps,
  parallel,
  ...
)

Value

the estimated penalty parameter

Arguments

data: a data frame containing a response, binary treatment indicators, and covariates.
Trt: a string specifying the name of the column of data contains the treatment indicators.
Y: a string specifying the name of the column of data contains the response.
zbar: the estimated marginal treatment effect
step1: character strings specifying the Step 1 model. Supports either "lasso", "mars", "randomforest", or "superlearner".
step2: a character string specifying the Step 2 model. Supports "lasso", "rtree", "classtree", or "ctree".
threshold: for "step2 = 'classtree'" only. The value against which to test if the estimated individual treatment effect from Step 1 is higher (TRUE) or lower (FALSE).
alpha0: the nominal Type I error rate.
p_reps: the number of permutations to run.
parallel: Should the loop over replications be parallelized? If FALSE, then no, if TRUE, then yes. Note that running in parallel requires a parallel backend that must be registered before performing the computation. See the foreach documentation for more details.
...: additional arguments to the Step 1 model call.