plot_coverage: Plot coverage probability of posterior credible intervals

In ImperialCollegeLondon/epidemia: Modeling of Epidemics using Hierarchical Bayesian Models

Description

Plots histograms showing empirical coverage of credible intervals specified using 'levels'. Can bucket by time period, by group, by whether the observation is new (not used in fitting).

Usage

plot_coverage(
  object,
  type,
  newdata = NULL,
  groups = NULL,
  levels = c(50, 95),
  period = NULL,
  by_group = FALSE,
  by_unseen = FALSE
)

Arguments

object	A fitted model object returned by epim. See epimodel-objects.
type	A string specifying the name of the observations to plot. This should match one of the names of the response variables in the obs argument used int the call to epim.
newdata	If provided, the original data used in object is overridden. Useful for forecasting
groups	Either NULL or a character vector specifying the groups to plot for. Default is NULL, which plots all modeled groups.
levels	A numeric vector defining the levels of the plotted credible intervals.
period	Buckets computed empirical probabilities into time periods if specified.
by_group	Plot coverage for each group individually
by_unseen	Plot coverage separately for seen and unseen observations. Observations are 'seen' if they were used for fitting.

Value

A ggplot object.

ImperialCollegeLondon/epidemia documentation built on June 26, 2021, 7:40 a.m.