Skip to contents

Compose a diagnostics dashboard for a fitted object

Source: R/plot-dashboard.R
plot_dashboard.Rd

A patchwork composition of the estimate panel with the weight and overlap diagnostics, annotated with the fit's metadata (n, m, k, s with provenance, distance, and jackknife type). CATE dashboards use estimates, overlap mass, local propensity, and cumulative weight share; nonparametric-regression dashboards use estimates, the diagnostics facets, and the weight decay and concentration views. Panels whose inputs are unavailable are substituted (the diagnostics facets take the place of a missing overlap panel) or dropped rather than failing the dashboard.

Usage

plot_dashboard(object, ...)

# Default S3 method
plot_dashboard(object, ...)

# S3 method for class 'cate_result'
plot_dashboard(object, ...)

# S3 method for class 'np_result'
plot_dashboard(object, ...)

Arguments

object

A fitted result object (cate_result or np_result).

...

Passed on to plot_estimates() for the estimates panel.

Value

A patchwork object (or a named list of ggplots when patchwork is unavailable).

Details

When patchwork is not installed the panels are returned as a named list of ggplot objects with a message.

Examples

if (requireNamespace("ggplot2", quietly = TRUE) &&
    requireNamespace("patchwork", quietly = TRUE)) {
  set.seed(1)
  n <- 100
  x_mat <- matrix(rnorm(n * 2), n, 2)
  y_vec <- x_mat[, 1] + rnorm(n, sd = 0.3)
  fit <- np_dnn(
    X = x_mat, X0 = cbind(seq(-1, 1, length.out = 6), 0), Y = y_vec,
    k = 1L, s = 12L, return = c("W", "dx")
  )
  plot_dashboard(fit)
}