boot_dec_quantile: Bootstrap of the quantile decomposition procedure.

In gibonet/decr: Tools to Perform Decompositions of Differences Between Two Groups

Bootstrap of the quantile decomposition procedure.

Description

Bootstrap estimates of dec_quantile and dec_. This serves to estimate for example confidence intervals of each component of the decomposition.

Usage

boot_dec_quantile(
  data,
  treatment,
  variables,
  y,
  weights = NULL,
  R = 10,
  probs = 0.5,
  counterfactual = c("AB", "BA"),
  ...
)

Arguments

data	data frame
treatment	column name of the binary variable
variables	character vector of the variables' names, for which the common support has to be established
y	name of the outcome variable for which you want to make the decomposition
weights	name of the weight variable (sample weights). If NULL (default value) it uses equal weights for all observations, adding a column of ones
R	The number of bootstrap replicates. This will be a single positive integer.
probs	numeric vector of length one with the desired quantile level (should be between 0 and 1).
counterfactual	"AB" or "BA". "AB" means that we want to estimate the counterfactual (wage) of group A, as if their characteristics were distributed as in group B. "BA" is the opposite (characteristics of group B are balanced to those of group A).
...	other arguments passed on to boot.

Examples

## Not run: 
data(invented_wages)
b01 <- boot_dec_quantile(data = invented_wages, treatment = "gender",
                        variables = c("sector", "education"), y = "wage",
                        R = 500, probs = 0.5)

## End(Not run)

gibonet/decr documentation built on Jan. 5, 2024, 7:26 a.m.