Observation streams for the Gaussian-approximation likelihood

ga_incidence() describes counts of an event over each observation interval (new infections, new deaths, new hospital admissions); ga_prevalence() describes counts of a compartment at the end of each interval. Reported counts are modelled as a binomial thinning of the latent count with probability prob plus Gaussian measurement error with variance noise times the latent count, approximated by a Gaussian with the same mean and variance.

Usage

ga_incidence(event, prob = 1, noise = 0)

ga_prevalence(compartment, prob = 1, noise = 0)

Arguments

event

Name of an event of the model.

prob

Name of the reporting-probability parameter, or a number in (0, 1].

noise

Name of the measurement-error parameter, or a non-negative number.

compartment

Name of a compartment of the model.

Value

An object of class ga_obs.