Skip to contents

Run a finite sample simulation by permuting treatment assignment labels within each block.

Source: R/rerandomize_simulation_code.R
run_rerandomize_simulation.Rd

Before starting simulation, put data in canonical form, imputing missing values and so forth as needed. Call compare_methods repeatidly on the resulting treatment-permuted data.

Usage

run_rerandomize_simulation(
  formula,
  data,
  R = 100,
  control_formula = NULL,
  summarize_results = FALSE,
  include_empirical = TRUE,
  warn_missing = TRUE,
  patch_data = TRUE,
  parallel = FALSE,
  ...
)

Arguments

formula

A formula object specifying the model to estimate.

data

A data frame containing the data to analyze.

R

The number of simulations to run.

control_formula

A formula object specifying the control variables; passed to compare_methods().

summarize_results

If TRUE, summarize the results of the compare_methods call, FALSE return raw estimates.

include_empirical

If TRUE, include the empirical results as one of the "simulation replicates". This will be included in any summarization, so be warned.

warn_missing

If TRUE, warn when rows are dropped or covariates are imputed while canonicalizing/patching the data.

patch_data

If TRUE, impute missing covariates via `patch_data_set()` before simulating.

parallel

If TRUE, run simulations in parallel.

...

Additional arguments to pass to compare_methods.