dfl_decompose_estimate {ddecompose}

Estimate the DFL reweighting decomposition

Description

This function performs the DFL decomposition. It derives the reweighting factors, estimates the distributional statistics and calculates the decomposition terms.

Usage

dfl_decompose_estimate(
  formula,
  dep_var,
  data_used,
  weights,
  group_variable,
  reference_group,
  method,
  estimate_statistics,
  statistics,
  probs,
  custom_statistic_function,
  right_to_left,
  trimming,
  trimming_threshold,
  return_model,
  estimate_normalized_difference,
  ...
)

Arguments

formula

formula object

dep_var

dependent variable

data_used

data.frame with data used for estimation

weights

weights variable

group_variable

group variable

reference_group

reference_group to be reweighted

method

method used to estimate conditional probabilities

estimate_statistics

boolean: if TRUE (default), then distributional statistics are estimated and the decomposition is performed. If FALSE, the function only returns the fitted inverse propensity weights.

statistics

a character vector that defines the distributional statistics for which the decomposition is performed.

probs

a vector of length 1 or more with the probabilities of the quantiles to be estimated.

custom_statistic_function

a function estimating a custom distributional statistic that will be decomposed.

right_to_left

determines the direction of a sequential decomposition.

trimming

boolean: If TRUE, observations with dominant reweighting factor values are trimmed according to rule of Huber, Lechner, and Wunsch (2013).

trimming_threshold

numeric: threshold defining the maximal accepted relative weight of the reweighting factor value (i.e., inverse probability weight) of a single observation. If NULL, the threshold is set to sqrt(N)/N, where N is the number of observations in the reference group.

return_model

boolean: If TRUE (default), the object(s) of the model fit(s) used to predict the conditional probabilities for the reweighting factor(s) are returned.

estimate_normalized_difference

boolean: If TRUE (default), the normalized differences between the covariate means of the comparison group and the reweighted reference group are calculated.

...

other parameters passed to the function estimating the conditional probabilities.