Skip to contents

ROC Curve Geom for Causal Inference

Source: R/geom_roc.R
geom_roc.Rd

A ggplot2 geom for plotting ROC curves with optional weighting. Emphasizes the balance interpretation where AUC around 0.5 indicates good balance.

Usage

geom_roc(
  mapping = NULL,
  data = NULL,
  stat = "roc",
  position = "identity",
  na.rm = TRUE,
  show.legend = NA,
  inherit.aes = TRUE,
  linewidth = 0.5,
  .focal_level = NULL,
  ...
)

Arguments

mapping

Set of aesthetic mappings. Must include estimate (propensity scores/predictions) and exposure (treatment/outcome variable). If specified, inherits from the plot.

data

Data frame to use. If not specified, inherits from the plot.

stat

Statistical transformation to use. Default is "roc".

position

Position adjustment. Default is "identity".

na.rm

If FALSE, the default, missing values are removed with a warning. If TRUE, missing values are silently removed.

show.legend

Logical. Should this layer be included in the legends? NA, the default, includes if any aesthetics are mapped.

inherit.aes

If FALSE, overrides the default aesthetics, rather than combining with them.

linewidth

Width of the ROC curve line. Default is 0.5.

.focal_level

The level of the exposure aesthetic to treat as the event. Must be a level the data actually takes; a declared factor level that no observation takes is not accepted. If NULL (default), the last observed level is used, which is the maximum value for numeric exposures.

...

Other arguments passed on to layer().

Value

A ggplot2 layer.

Details

A curve compares two exposure levels, so each curve is drawn from the rows that share an aesthetic signature, with the exposure level excluded from that signature. Mapping group explicitly makes each group a curve of its own, which keeps long data holding several weighting schemes from being pooled into a single curve. A group that holds only one observed exposure level is dropped with a warning, and the remaining curves are still drawn.

See also

check_model_auc() for computing AUC values, stat_roc() for the underlying stat

Other ggplot2 functions: geom_calibration(), geom_ecdf(), geom_mirror_density(), geom_mirror_histogram(), geom_qq2()

Examples

# Basic usage
library(ggplot2)
ggplot(nhefs_weights, aes(estimate = .fitted, exposure = qsmk)) +
  geom_roc() +
  geom_abline(intercept = 0, slope = 1, linetype = "dashed")


# With grouping by weight
long_data <- tidyr::pivot_longer(
  nhefs_weights,
  cols = c(w_ate, w_att),
  names_to = "weight_type",
  values_to = "weight"
)
#> Warning: Converting psw to numeric: incompatible estimands 'ate' and 'att'
#> ℹ Metadata cannot be preserved when combining incompatible objects
#> ℹ Use identical objects or explicitly cast to numeric to avoid this warning

ggplot(long_data, aes(estimate = .fitted, exposure = qsmk, weight = weight)) +
  geom_roc(aes(color = weight_type)) +
  geom_abline(intercept = 0, slope = 1, linetype = "dashed")