Custom function to visualize the balance between treatment and control groups

Usage

balance_scatter_custom(
  pm_result_list,
  panel.data,
  set.names = NULL,
  show.legend = TRUE,
  legend.title = "Type",
  legend.position = "right",
  xlim = c(0, 0.8),
  ylim = c(0, 0.8),
  main = "Standardized Mean Difference of Covariates",
  pchs = NULL,
  dot.size = NULL,
  covariates,
  x.axis.label = "Before Refinement",
  y.axis.label = "After Refinement",
  theme_use = causalverse::ama_theme(),
  ...
)

Arguments

pm_result_list: List of PanelMatch result objects (output of PanelMatch::PanelMatch()).
panel.data: A PanelData object created by PanelMatch::PanelData().
set.names: Vector of names for matched sets. Defaults to NULL.
show.legend: Boolean to determine if legend should be shown. Defaults to TRUE.
legend.title: Legend title. Defaults to "Type".
legend.position: Position of legend. Defaults to "right".
xlim: Vector defining x-axis limits. Defaults to c(0, 0.8).
ylim: Vector defining y-axis limits. Defaults to c(0, 0.8).
main: Main title for the plot. Defaults to "Standardized Mean Difference of Covariates".
pchs: Plot characters. Defaults to NULL.
dot.size: Size of dots in the scatter plot. Defaults to NULL.
covariates: Covariates for calculating balance.
x.axis.label: x-axis label. Defaults to "Before Refinement".
y.axis.label: y-axis label. Defaults to "After Refinement".
theme_use: Custom theme that follows ggplots2. Defaults to causalverse::ama_theme().
...: Additional arguments passed to the labs() function

Value

ggplot object

Examples

if (FALSE) { # \dontrun{
library(PanelMatch)
pd <- PanelData(panel.data = dem, unit.id = "wbcode2",
                time.id = "year", treatment = "dem", outcome = "y")

# Maha 4-year lag, up to 5 matches
PM.results.5m <- PanelMatch(
   panel.data = pd,
   lag = 4,
   refinement.method = "mahalanobis",
   match.missing = TRUE,
   covs.formula = ~ I(lag(tradewb, 1:4)) + I(lag(y, 1:4)),
   size.match = 5,
   qoi = "att",
   lead = 0:4,
   forbid.treatment.reversal = FALSE,
   use.diagonal.variance.matrix = TRUE
)

# Maha 4-year lag, up to 10 matches
PM.results.10m <- PanelMatch(
   panel.data = pd,
   lag = 4,
   refinement.method = "mahalanobis",
   match.missing = TRUE,
   covs.formula = ~ I(lag(tradewb, 1:4)) + I(lag(y, 1:4)),
   size.match = 10,
   qoi = "att",
   lead = 0:4,
   forbid.treatment.reversal = FALSE,
   use.diagonal.variance.matrix = TRUE
)

# Using the function
balance_scatter_custom(
   pm_result_list = list(PM.results.5m, PM.results.10m),
   panel.data = pd,
   set.names = c("Maha 4 Lag 5 Matches", "Maha 4 Lag 10 Matches"),
   covariates = c("y", "tradewb")
)
} # }