conditional_cases_logp: Calculates (unnormalised) log-probability of a true case...

In ben18785/incidenceinflation: Estimate R0 Given Case Reporting Delays

Calculates (unnormalised) log-probability of a true case count pertaining to cases arising on a particular onset day

Description

There are two components to this:

p(cases_true_t|data, Rt, reporting_params, serial_params) \propto p(data|cases_true, reporting_params) p(cases_true_t|cases_true_t_1, cases_true_t_2, ..., Rt, serial_params)

Usage

conditional_cases_logp(
  cases_true,
  observation_df,
  cases_history,
  Rt,
  day_onset,
  serial_parameters,
  reporting_parameters,
  kappa = NULL,
  is_negative_binomial = FALSE
)

Arguments

cases_true	true case count originating on day_onset
observation_df	a two-column data frame with columns 'time_reported' and 'cases_reported'
cases_history	a vector containing history of cases arranged from recent to past
Rt	effective reproduction number
day_onset	day when case originates
serial_parameters	named list of 'mean' and 'sd' of gamma distribution characterising the serial interval distribution
reporting_parameters	named list of 'mean' and 'sd' of gamma distribution characterising the reporting delay distribution
kappa	overdispersion parameter
is_negative_binomial	if negative-binomial renewal model specified (defaults to FALSE)

Value

an (unnormalised) log-probability or vector of such log-probabilities

ben18785/incidenceinflation documentation built on Feb. 8, 2024, 2:36 a.m.