Estimate Severity From A Line List — estimate

Fit a severity estimate model accounting for under reporting of asymptomatic and mildly symptomatic cases.

Usage

estimate_severity(
  linelist,
  population,
  surveillance,
  outcome,
  time_period = character(),
  strata = character(),
  population_value = utils::tail(names(population), n = 1L),
  surveillance_reference = NULL,
  outcome_reference = NULL,
  time_period_reference = NULL,
  strata_reference = NULL,
  hazard_std = 3,
  degrees_of_freedom = 1L,
  active_prior = c(alpha = 1, beta = 25),
  passive_asymptomatic_prior = c(alpha = 1, beta = 2),
  passive_symptomatic_prior = c(alpha = 25, beta = 1),
  ...
)

Arguments

linelist: A data.frame (or data.frame extending object like a tibble) of the line list data.
population: A data.frame (or data.frame extending object like a tibble) of the population data.
surveillance: A character of columns describing the surveillance methods. The values in this column must be coercible to 'Active' and 'Passive'.
outcome: A character of columns describing the outcome of the line list entry. The values in this column must be coercible to 'Asymptomatic', 'Death', and 'Symptomatic'.
time_period: A character of columns describing the time period such as 'week' or 'day'
strata: A character of columns describing the attributes to stratify the data on.
population_value: A unit length character corresponding to the column in the population data.frame (or data.frame like object) for the population value.
surveillance_reference: Either NULL, a data.frame, or a vector of values to use as a reference for the surveillance dimension. If a vector then it is expected that surveillance is unit length.
outcome_reference: Either NULL, a data.frame, or a vector of values to use as a reference for the outcome dimension. If a vector then it is expected that outcome is unit length.
time_period_reference: Either NULL, a data.frame, or a vector of values to use as a reference for the time period dimension. If a vector then it is expected that time_period is unit length.
strata_reference: Either NULL, a data.frame, or a vector of values to use as a reference for the strata dimension. If a vector then it is expected that strata is unit length.
hazard_std: A single length numeric greater than zero for the standard deviation to use in the community hazard prior distribution.
degrees_of_freedom: A single length integer greater than zero for the degrees of freedom to use in the model.
active_prior: The parameters for the prior beta distribution for the active detection probability. Can be specified as 'alpha'/'beta', 'mean'/'var', 'mean'/'sd', or 'mean'/'concentration'.
passive_asymptomatic_prior: The parameters for the prior beta distribution for the passive asymptomatic detection probability. Can be specified as 'alpha'/'beta', 'mean'/'var', 'mean'/'sd', or 'mean'/'concentration'.
passive_symptomatic_prior: The parameters for the prior beta distribution for the passive symptomatic detection probability. Can be specified as 'alpha'/'beta', 'mean'/'var', 'mean'/'sd', or 'mean'/'concentration'.
...: Further optional args that are eventually given to rstan::sampling() related to fitting.

Value

A SeverityEstimateFit S4 object.