autoplot method for summary.simsum objects
# S3 method for summary.simsum
autoplot(
object,
type = "forest",
stats = "nsim",
target = NULL,
fitted = TRUE,
scales = "fixed",
top = TRUE,
density.legend = TRUE,
zoom = 1,
zip_ci_colours = "yellow",
...
)A ggplot object.
An object of class summary.simsum.
The type of the plot to be produced. Possible choices are: forest, lolly, zip, est, se, est_ba, se_ba, est_ridge, se_ridge, est_density, se_density, est_hex, se_hex, heat, nlp, with forest being the default.
Summary statistic to plot, defaults to nsim (the number of replications with non-missing point estimates/SEs).
See summary.simsum() for further details on supported summary statistics.
Target of summary statistic, e.g. 0 for bias. Defaults to NULL, in which case target will be inferred.
Superimpose a fitted regression line, useful when type = (est, se, est_ba, se_ba, est_density, se_density, est_hex, se_hex). Defaults to TRUE.
Should scales be fixed (fixed, the default), free (free), or free in one dimension (free_x, free_y)?
Should the legend for a nested loop plot be on the top side of the plot? Defaults to TRUE.
Should the legend for density and hexbin plots be included? Defaults to TRUE.
A numeric value between 0 and 1 signalling that a zip plot should zoom on the top x% of the plot (to ease interpretation). Defaults to 1, where the whole zip plot is displayed.
A string with (1) a hex code to use for plotting coverage probability and its Monte Carlo confidence intervals (the default, with value zip_ci_colours = "yellow"), (2) a string vector of two hex codes denoting optimal coverage (first element) and over/under coverage (second element) or (3) a vector of three hex codes denoting optimal coverage (first), undercoverage (second), and overcoverage (third).
Not used.
data("MIsim", package = "rsimsum")
s <- rsimsum::simsum(
data = MIsim, estvarname = "b", true = 0.5, se = "se",
methodvar = "method", x = TRUE
)
ss <- summary(s)
library(ggplot2)
autoplot(ss)
autoplot(ss, type = "lolly")
Run the code above in your browser using DataLab