propensity_score: Calculate propensity scores.

In jackcollison/causality: Tools for Performing Causal Inference with Observational Data

Description

Function that calculates propensity scores with an option to check calibration and overlap assumptions

Usage

propensity_score(
  data,
  x,
  y,
  w,
  subset = FALSE,
  model = "logistic",
  trim = FALSE,
  plot = TRUE,
  ...
)

Arguments

data	a dataframe object containing the variables and values.
x	a list of character vectors specifying variables to be included in the model (columns in the data). If unspecified, then it is assumed to be all columns in the data besides y and w.
y	a character vector specifying the response variable.
w	a character vector specifying the treatment variable. Treatment must be specified as 0 and 1 or TRUE and FALSE.
subset	an optional vector specifying a subset of observations to be used in the fitting process.
model	a character string naming the model used (one of "logistic", "lasso", or "causal forest").
trim	a list of two numeric elements describing cutoffs at which to trim propensity scores.
plot	logical; if TRUE then produces histogram of propensity scores to check that the overlap assumption is satisfied and a Q-Q plot to check the calibration.
...	additional arguments to be passed to the model fitting functions.

Value

a vector of propensity scores where the indices of the vector correspond to the indices in data.

References

Imbens, Guido, and Donald Rubin. Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction. Cambridge University Press, 2015.

Examples

data(lalonde)
logit_propensities <- propensity_score(lalonde, y = "re78", w = "treat", model = "logistic")
lasso_propensities <- propensity_score(lalonde, y = "re78", w = "treat", model = "lasso")
cf_propensities <- propensity_score(lalonde, y = "re78", w = "treat", model = "causal forest")

jackcollison/causality documentation built on Dec. 20, 2021, 8:05 p.m.