net_simulate: Network simulation wrapper

In AIforGoodSimulator/network-model-R-package: CNS(CampNetworkSimulator)

Description

This function simulates a SEIQHRF epidemic on top of a network of n individuals with a specific list of parameters as provided. S - Susceptible E - Exposed (also Infected) I - Infected Q - Quarantined (also Infected) H - Requiring hospitalisation (also Infected) R - Recovered F - Fatality (due to infection)

Usage

net_simulate(
  n = 100,
  nsims = 3,
  nsims_dx = 100,
  nsteps = 90,
  prop.isobox = 0.43,
  iso.capacity = 10,
  tent.capacity = 4,
  verbose = F,
  cores = 1,
  d.rate = 0,
  external.contacts = 4,
  act.rate.se = 10,
  inf.prob.se = 0.02,
  act.rate.si = 10,
  inf.prob.si = 0.05,
  act.rate.sq = 2.5,
  inf.prob.sq = 0.05,
  ei.rate = 1/10,
  iq.rate = 1/30,
  ih.rate = 1/100,
  qh.rate = 1/100,
  hr.rate = 1/15,
  qr.rate = 1/20,
  ir.rate = 1/20,
  hf.rate = 1/50,
  hf.rate.overcap = 1/25,
  hosp.cap = 5,
  hosp.tcoeff = 0.5,
  a.rate = 0,
  di.rate = d.rate,
  ds.rate = d.rate,
  dr.rate = d.rate,
  i.num = 1,
  r.num = 0,
  e.num = 0,
  s.num = n - 1,
  f.num = 0,
  h.num = 0,
  q.num = 0,
  agedistribution = CampNetworkSimulator::agedistribution,
  ratesbyage = CampNetworkSimulator::ratesbyage
)

Arguments

n	numbers of individuals, a random sample of this size will be taken from the Camp's population
nsims	number of network simulations (default: 3) ~ simulation sample size
nsims_dx	number of network simulation for network diagnostics (default: 100)
nsteps	number of timesteps (days, hours, weeks, arbitrary choice), the rates need to be defined in units relative to this metric (e.g number of cases/day or /hour or /week)
prop.isobox	Proportion of individuals in iso boxes, in this first implementation only two kinds of housing units are considered (isoboxes and tent) – can (and should) be changed in the future
iso.capacity	Capacity of iso boxes
tent.capacity	Capacity of tents
verbose	whether function should be verbose in its command line output
cores	number of cores to be used (default: number of available cores - 1)
d.rate	departure rate (includes external deaths and actual departures) – if set to 0 no invidual is consider to leave or die from causes different than the epidemic agent
external.contacts	Average number of external contacts that individuals make every day (external meaning from outside their tent)
act.rate.se	Act rate (contact rate) between susceptible and exposed individuals – default: 10 interactions of S-to-E/timestep/person
inf.prob.se	infection probability of susceptible individuals upon contact with exposed individuals, these value is often considered to be lower than that upon contact with infected individuals (though literature needs to be checked for this) – default: 0.02 (Out of every 100 contacts between S and E we expect 2 infections)
act.rate.si	Act rate (contact rate) between susceptible and infected individuals – default: 10 interactions/timestep /person
inf.prob.si	infection probability of susceptible individuals upon contact with susceptible individuals – default: 0.05
act.rate.sq	Act rate (contact rate) between susceptible and quarantined individuals – default: 2.5 interactions/timestep/person
inf.prob.sq	infection probability of susceptible individuals upon contact with quarantined individuals – default: 0.05
ei.rate	Rate (can be seen as flow rate or probability) between exposed and infected individuals, can take the value of a vector of length nsteps to simulate timevarying scenarios
iq.rate	Rate between infected and quarantined states (~chances of quarantining when infected)
ih.rate	Rate between infected and hospitalised states (~chances of requiring hospitalisation when infected)
qh.rate	Rate between quarantined and hospitalised states (~ chances of requiring hospitalisation when quarantining)
hr.rate	Rate between requiring hospitalisation and recovered states (~chances of recovering when requiring hospitalisation)
qr.rate	Rate between quarantined and recovered states (~chances of recovering when in quarantined state)
ir.rate	Rate between infected and recovered states (~chances of recovering when in infected state)
hf.rate	Rate between hospitalised and fatality state (~chances of dying from the infectious agent given requiring hospitalisation)
hf.rate.overcap	Same as hf.rate but when the hospital is at full capacity (As per hosp.cap)
hosp.cap	Hospital capacity
hosp.tcoeff	Increase in the hr.rate for every timestep spent in the hospital
a.rate	arrival rate -> if set to 0, inflow of people is not considered (i.e. no births, no immigration)
di.rate	departure rate for infected individuals (defaults to d.rate)
ds.rate	departure rate for susceptible individuals (defaults to d.rate)
dr.rate	departure rate for susceptible individuals (defaults to d.rate)
i.num	Initial number of infected individuals (defanetults to 1)
r.num	Initial number of recovered individuals (defaults to 0)
e.num	Initial number of exposed individuals (default to 0)
s.num	Initial number of susceptible individuals (defaults to n-1)
f.num	Initial number of fatalities (defaults to 0)
h.num	Initial number of individuals requiring hospitalization (defaults to 0)
q.num	Initial number of quarantined individuals (defaults to 0)
agedistribution	vector containing the age distribution of individuals in the camp. By default this will be loaded internally, see CampNetworkSimulator::agedistribution
ratesbyage	dataframe containing the hospitalisation rate and fatality rate by age, by default CampNetworkSimulator::ratesbyage is used

Examples

net_simulate(n= 100, nsims = 1, nsteps = 10)

AIforGoodSimulator/network-model-R-package documentation built on Oct. 3, 2020, 10:08 a.m.