Universal Coefficient Plot — coef

Creates a publication-ready coefficient (forest) plot from any model or list of models. Supports lm, glm, fixest, ivreg, and tidy data frames.

Usage

coef_plot(
  models,
  terms = NULL,
  term_labels = NULL,
  model_labels = NULL,
  sort_by = c("estimate", "name", "none"),
  add_zero_line = TRUE,
  color_by = c("model", "significance"),
  conf_level = 0.95,
  dodge_width = 0.4,
  title = NULL,
  xlab = "Estimate (95 pct CI)"
)

Arguments

models: A model object, a named list of model objects, or a tidy data frame with columns term, estimate, conf.low, conf.high (broom-style) or ci_lower, ci_upper (causalverse-style).
terms: Character vector. Terms to include. If NULL (default), all non-intercept terms are shown.
term_labels: Named character vector mapping original term names to display labels, e.g. c(am = "Transmission (auto)").
model_labels: Character vector. Display names for each model, used in the legend. Defaults to list element names or "Model".
sort_by: Character. Sort terms by "estimate" (default), "name", or "none".
add_zero_line: Logical. Add a vertical reference line at zero. Default TRUE.
color_by: Character. Color points by "model" (default) or "significance" (significant at 5 percent in a different color).
conf_level: Numeric. Confidence level for intervals extracted from model objects. Default 0.95.
dodge_width: Numeric. Horizontal offset between multiple models. Default 0.4.
title: Character or NULL. Plot title.
xlab: Character. X-axis label. Default "Estimate (95 pct CI)".

Value

A ggplot2 object.

Examples

# Single lm model
mod <- lm(mpg ~ am + wt + hp, data = mtcars)
coef_plot(mod, terms = c("am", "wt", "hp"),
          term_labels = c(am = "Transmission (auto)",
                          wt = "Weight", hp = "Horsepower"))


if (FALSE) { # \dontrun{
# Multiple models
mod2 <- lm(mpg ~ am + wt + hp + disp, data = mtcars)
coef_plot(
  list(Parsimonious = mod, Full = mod2),
  terms = c("am", "wt", "hp"),
  color_by = "significance"
)
} # }