outbreaker_chains: Basic methods for processing outbreaker results
In o2geosocial: Reconstruction of Transmission Chains from Surveillance Data

print.outbreaker_chains

R Documentation

Basic methods for processing outbreaker results

Description

Several methods are defined for instances of the class outbreaker_chains, returned by outbreaker, including: print, plot

Usage

## S3 method for class 'outbreaker_chains'
print(x, n_row = 3, n_col = 8, type = "chain", ...)

## S3 method for class 'outbreaker_chains'
plot(
  x,
  y = "post",
  type = c("trace", "hist", "density", "cluster", "alpha", "t_inf", "kappa", "network"),
  burnin = 0,
  min_support = 0.1,
  labels = NULL,
  group_cluster = NULL,
  ...
)

## S3 method for class 'outbreaker_chains'
summary(object, burnin = 0, group_cluster = NULL, ...)

Arguments

`x`	an `outbreaker_chains` object as returned by `outbreaker`.
`n_row`	the number of rows to display in head and tail; defaults to 3.
`n_col`	the number of columns to display; defaults to 8.
`type`	a character string indicating the kind of plot to be used (see details)
`...`	further arguments to be passed to other methods
`y`	a character string indicating which element of an `outbreaker_chains` object to plot
`burnin`	the number of iterations to be discarded as burnin
`min_support`	a number between 0 and 1 indicating the minimum support of ancestries to be plotted; only used if 'type' is 'network'
`labels`	a vector of length N indicating the case labels (must be provided in the same order used for dates of symptom onset)
`group_cluster`	a numeric `vector` indicating the breaks to aggregate the cluster size distribution.
`object`	an `outbreaker_chains` object as returned by `outbreaker`.

Details

type indicates the type of graphic to plot:

trace to visualise MCMC traces for parameters or augmented data (plots the log-likelihood by default)
hist to plot histograms of quantitative values
density to plot kernel density estimations of quantitative values
alpha to visualise the posterior frequency of ancestries
network to visualise the transmission tree; note that this opens up an interactive plot and requires a web browser with Javascript enabled; the argument 'min_support' is useful to select only the most supported ancestries and avoid displaying too many links
kappa to visualise the distributions generations between cases and their ancestor/infector
cluster to visualise the cluster size distribution, grouped by the value in group_cluster

Value

The form of the value returned by plot depends on the type. If the type is set as network, plot returns a visNetwork object containing the details of the inferred transmission trees. Otherwise, it returns a ggplot object containing the elements of the plot.

The function summary returns a list containing 9 elements:

step: contains the first and last values of the iteration number; the interval between each iteration retained for the output (defined by the parameter sample_every in create_config), and the number of iterations in the output,
post: contains the minimum, maximum, mean, median and quartiles of the posterior distribution.
like: contains the minimum, maximum, mean, median and quartiles of the likelihood distribution.
prior: contains the minimum, maximum, mean, median and quartiles of the prior distribution.
pi: contains the minimum, maximum, mean, median and quartiles of the conditional report ratio.
a: contains the minimum, maximum, mean, median and quartiles of the spatial parameter a.
b: contains the minimum, maximum, mean, median and quartiles of the spatial parameter b.
tree: a data.frame that contains the most likely infector, the infection date, and the number of missing generations of each case. It also contains the support of the most likely branch (i.e. the proportion of iterations where the infector of a case is its most likely infector), and import, the proportion of iteration where the case was classified as an importation.
cluster: a data frame listing the minimum, maximum, median, mean and quartile of the cluster size distribution.