transport: Balancing Weights for Transporting Observational Results

In kevjosey/cbal: Covariate Balancing Weights using Generalized Projections of Bregman Distances

Balancing Weights for Transporting Observational Results

Description

The transport_ATE() function find balancing weights for transporting estimates. Transport weights requires the complete individual-level data (response, treatment assignment, and covariates) for all units in the observed sample but only the individual-level covariate data from the target sample.

Usage

transport_ATE(
  S,
  X,
  Y1,
  Z1,
  base_weights = NULL,
  optim_ctrl = list(maxit = 500, reltol = 1e-10),
  ...
)

Arguments

S	the binary vector of sample indicators.
X	the balance functions to be contrained.
Y1	the observed responses for S = 1.
Z1	the binary treatment assignment for S = 1.
base_weights	a vector of optional base weights with length equal to the number of rows in X.
optim_ctrl	a list of arguments that will be passed to optim().
...	additional arguments.

References

Josey KP, Berkowitz SA, Ghosh D, Raghavan S (2020a). "Transporting Experimental Results with Entropy Balancing." arXiv:2002.07899 [stat].

Josey KP, Yang F, Ghosh D, Raghavan S (2020b). "A Calibration Approach to Transportability and Data-Fusion with Observational Data." arXiv:2008.06615 [stat].

kevjosey/cbal documentation built on July 22, 2023, 11:04 a.m.