Skip to contents

Link coarseDataTools and EpiEstim

Source: R/coarse2estim.R
coarse2estim.Rd

coarse2estim Transforms outputs of dic.fit.mcmc to right format for input into estimate_R

Usage

coarse2estim(x = NULL, dist = x@dist, samples = x@samples, thin = 10)

Arguments

x

An object generated by function dic.fit.mcmc, containing posterior estimates of the serial interval distribution.

dist

The parametric distribution used when estimating the serial interval. Should be one of "gamma", "weibull", "lognormal", "gamma_offset_1", "weibull_offset_1", or "lognormal_offset_1". Note the different naming convention compared to dic.fit.mcmc. The distribution may also be specified using the abbreviated forms "G", "W", "L", "off1G", "off1W", and "off1L" as used in dic.fit.mcmc. However, we recommend using the full names to avoid confusion, and a warning will be issued if the abbreviated forms are used. If not present, computed automatically from x.

samples

A dataframe containing the posterior samples of serial interval parameters corresponding to the parametric choice specified in dist. If not present, computed automatically from x.

thin

A positive integer corresponding to thinning parameter; of the posterior sample of serial interval distributions in x, only 1 in thin will be kept, the rest will be discarded.

Value

A list with two elements:

  • si_sample: a matrix where each column gives one distribution of the serial interval to be explored, obtained from x by thinning the MCMC chain.

  • si_parametric_distr: the parametric distribution used when estimating the serial interval stored in x.

See also

estimate_R()

Author

The Hackout3 Parameter Estimation team.

Examples

if (FALSE) { # \dontrun{
## Note the following examples use an MCMC routine
## to estimate the serial interval distribution from data,
## so they may take a few minutes to run

## load data on rotavirus
data("MockRotavirus")

## estimate the serial interval from data
SI.fit <- coarseDataTools::dic.fit.mcmc(
  dat = MockRotavirus$si_data,
  dist = "G",
  init.pars = init_mcmc_params(MockRotavirus$si_data, "G"),
  burnin = 1000,
  n.samples = 5000
)

## use coarse2estim to turn this in the right format for estimate_R
si_sample <- coarse2estim(SI.fit, thin = 10)$si_sample

## use estimate_R to estimate the reproduction number
## based on these estimates of the serial interval
R_si_from_sample <- estimate_R(
  MockRotavirus$incidence,
  method = "si_from_sample",
  si_sample = si_sample,
  config = make_config(list(n2 = 50))
)

plot(R_si_from_sample)
} # }