Skip to contents

Functions for plots that assess forecast calibration Get quantile-quantile plot

Source: R/plot_calibration.R
forecast_qq_plot.Rd

Adapted from MIT-licensed scoringutils::plot_quantile_coverage().

Usage

forecast_qq_plot(
  forecasts,
  model_z_order = NULL,
  linewidth = 2,
  reference_linecolor = "gray",
  reference_linewidth = 2,
  reference_linetype = "dashed",
  ...
)

Arguments

forecasts

df of granular (daily) quantile forecasts

model_z_order

z-order in which to overplot the individual models, ascending (so the last named model is plotted on top). If NULL (default), plot the models in the order they appear in forecasts.

linewidth

width parameter for the Q-Q lines for individual models. Default 2.

reference_linecolor

color parameter for the r eference y = x line. Default "gray".

reference_linewidth

width parameter for the reference y = x line. Default 2.

reference_linetype

linetype parameter for the reference y = x line. Default "dashed".

...

additional keyword arguments passed to ggplot2::geom_line() for the Q-Q lines for individual models.

Value

a ggplot object containing a plot of the proportion of data within each interval for each model.