Bayesian Compartmental Infection Models from Individual Outcomes

Fit Bayesian compartmental infection models (such as susceptible-infected-recovered and multi-compartment hidden Markov variants) from individual-level outcome data. Models are composed from infection-process and observation-process components and fit using Stan via 'rstan'.

hestia

{hestia} fits Bayesian compartmental infection models — such as susceptible-infected-recovered (SIR) and multi-compartment hidden Markov variants — from individual-level outcome data. Models are composed from infection-process and observation-process components and fit using Stan via {rstan}.

Installation

remotes::install_github("ACCIDDA/hestia", build = TRUE)

Requires R >= 4.1.0 and a C++ toolchain (for Stan model compilation).

Usage

Compose an infection-process model from transmit() (transmission between compartments) and progress() (within-host progression) steps:

library(hestia)

# A basic SIR model: S -> I transmission, I -> R recovery (rate fit from data)
inf_process <- make_infection_model(
  transmit(from = "S", to = "I"),
  progress(from = "I", to = "R", gamma = NA)
)

Pair the infection process with an observation model (make_observation_model()) and fit it with run_model(). See the vignettes for full worked examples, including a multi-compartment (symptomatic/asymptomatic) model.