Plot cross-sectional causal decomposition

Usage

# S3 method for class 'ineqx_causal_cross'
plot(
  x,
  type = "wibe",
  ci = FALSE,
  stats = NULL,
  trim = 0.995,
  share = FALSE,
  ...
)

Arguments

x

An ineqx_causal_cross object

type

Character, the plot type:

"wibe": Within/between treatment effects (bar chart).
"wibe.group": Group-level contributions to within/between.
"treat": Predicted treatment effect distributions by group.
"treat.params": Treatment effect parameters (beta, lambda) by group (bar chart).
"effect.prop": Proportional treatment effect on the mean by group (points): beta_g / mu0_g (identity) or exp(beta_g)-1 (log). Equal heights indicate a proportional (group-invariant) effect.
"outcome": Predicted outcome distributions (control vs treated) by group.
"outcome.params": Predicted means and SDs under control vs treatment. For simple-difference models, a bar chart per group. For DiD models, a netted-out line chart anchored at the pre-period treated level: the post-period gap between observed and counterfactual lines equals the DiD ATT.

ci

Whether to show confidence intervals. Accepts FALSE or "none" (no CIs), TRUE or "delta" (use stored SEs), "boot" (bootstrap with defaults), or boot_config() (bootstrap with custom settings). Default FALSE.

stats

Character vector. For type = "wibe": components to display. The bar plot recognizes the same canonical vocabulary as the longitudinal type = "wibe" method: "tau" (Total τ bar), "tau_b" (Between τ bar), "tau_w" (Within τ bar), "het_b", "cov_b", "rescale_b" (Between sub-components: heterogeneity, sorting, rescaling), and the within mirrors "het_w", "cov_w", "rescale_w". Single-segment bars are allowed. The legend labels "cov_*" as “Sorting” and "rescale_*" as “Rescaling”. Shorthands: "het_cov" = the four heterogeneity / sorting sub-components (no rescaling), "het_cov_b" = c("het_b","cov_b"), "het_cov_w" = c("het_w","cov_w"), "subs" = all six sub-components (het + cov + rescale, between and within), "subs_b" = c("het_b","cov_b","rescale_b"), "subs_w" = c("het_w","cov_w","rescale_w"). When tau and any sub-component appear at the same column, the bars are dodged. Default "tau" preserves the classic three-bar (Total/Between/Within) layout; c("tau","het_cov") preserves the dodged classic layout.

trim

Numeric between 0 and 1. For distribution plots ("treat", "outcome"): quantile at which to trim tails. Default 0.995 (trims 0.5% from each tail).

share

Logical. For type = "wibe", show the Between/Within split as shares of the total effect (%) rather than absolute units. Errors if the total effect is \(\approx 0\) (shares undefined). Default FALSE.

...

Additional arguments (currently unused)

Value

A ggplot2 object