EpiModel (version 0.25)

epiABM: Stochastic Agent-Based SI-SIS-SIR Models

Description

This function simulates an agent-based model of an infectious disease with random mixing in the population.

Usage

epiABM(type, s.num, i.num, r.num = 0, beta, cont, nu = 0,
    nsteps, nsims, verbose = TRUE)

Arguments

type

model type (choice of 'SI', 'SIR', or 'SIS')

s.num

number of initial susceptible in population

i.num

number of initial infected in population

r.num

number of initial recovered in poulation

beta

probability of infection per contact

cont

average number of contacts per person per unit time

nu

average rate of recovery (1/duration of disease)

nsteps

number of time steps to simulate over

nsims

number of simulations to run

verbose

print progress

Details

This function provides flexibility in model choice of a susceptible-infected (SI), susceptible-infected-recovered (SIR), or a susceptible-infected-susceptible (SIS) epidemic model. For an SI model, use the defaults of 0 for r.num, nu, and mu. For an SIR model, specify the recovery rate with the nu parameter, but leave mu to its 0 default. For an SIS model, specify the re-susceptiblity with the mu parameter, but leave the nu to its 0 default.

The model has two sources of stochasticity. First, the contact structure is random, with new partnerships selected at random from among the actors. The contact rate argument cont is a rate per person per timestep, and is transformed to the number of new partnerships per timestep by cont*(s.num+i.num)/2. Second, the probability of transmission within each partnership is based on a standard uniform random deviate falling below beta.

See Also

plotABM

Examples

# SI model
out <- epiABM(type='SI', s.num=500, i.num=1, beta=0.2, cont=0.25, nsteps=250, nsims=10)
par(mar=c(3,3,1,1), mgp=c(2,1,0))
plotABM(out, 'i.num', medline=FALSE)

# SIR model
out <- epiABM(type='SIR', s.num=500, i.num=1, beta=0.2, cont=0.25, nu=1/50, nsteps=250, nsims=10)
plotABM(out, 'i.num', medline=FALSE)

# SIS model
out <- epiABM(type='SIS', s.num=500, i.num=1, beta=0.2, cont=0.25, nu=1/50, nsteps=250, nsims=10)
plotABM(out, 'i.num', medline=FALSE)