balance_plot: Assess balance in covariate distribution graphically.

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

Description

Assess balance in covariate distributio and assess ability to adjust for differences in covariates by treatment status.

Usage

balance_plot(
  data,
  x,
  w,
  matched_indices = NULL,
  method = "Normalized difference",
  threshold = 0.2,
  colors = NULL,
  title = NULL
)

Arguments

data	a dataframe object containing the variables and values.
x	a list of character vectors specifying the covariates.
w	a character vector describing the treatment variable.
matched_indices	a data frame of two columns, one of treatment indices and one of matched (produced by propensity_match).
method	a character vector specifying the method for which to assess balance in covariate distributions across treatment and control groups.
threshold	the threshold to plot. Default is 0.2.
colors	an optional vector of two specifying the colors for the control and treatment groups, respectively.
title	an optional plot title.

Value

a ggplot object.

Examples

data(lalonde)

p <- propensity_score(lalonde, y = "re78", w = "treat")
matched_indices <- propensity_match(lalonde, w = "treat", p = p, max_distance = .00001)

vars <- names(lalonde)
covariates <- vars[!vars %in% c("re78", "treat")]

balance_plot(lalonde, x = covariates, w = "treat", matched_indices = matched_indices)

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