plot.S3sensitivity {BuyseTest}

Graphical Display for Sensitivity Analysis

Description

Display the statistic of interest across various threshold values, possibly with confidence intervals. Currently only works when varying thresholds relative to one or two variables.

Usage

## S3 method for class 'S3sensitivity'
plot(x, plot = TRUE, ...)

## S3 method for class 'S3sensitivity'
autoplot(
  object,
  col = NULL,
  ci = TRUE,
  band = TRUE,
  label = "Threshold for",
  position = NULL,
  size.line = 1,
  size.point = 1.75,
  size.ci = 0.5,
  alpha = 0.1,
  ...
)

Arguments

plot

[logical] should the graph be displayed in a graphical window

...

not used. For compatibility with the generic method.

object, x

output of the sensitivity method

col

[character vector] color used to identify the thresholds relative to a second variable.

ci

[logical] should the confidence intervals be displayed?

band

[logical] should the simulatenous confidence intervals be displayed?

label

[character] text used before the name of the variables in the legend.

position

relative position of the error bars for a given x value. Can for instance be position_dodge(width = 5).

size.line

[numeric] width of the line connecting the point estimates.

size.point

[numeric] size of the point representing the point estimates.

size.ci

[numeric] width of the lines representing the confidence intervals.

alpha

[numeric] transparency for the area representing the simultaneous confidence intervals.

Details

The autoplot and plot methods are very similar. The main difference is that the former returns a ggplot2 object whereas the later automatically display the figure in a graphical window and returns an (invible) list with the plot and the data.

Value

a ggplot2 object