DEN.spatial.ensemble: Run a multiple realisations of the spatial dengue model and...

In obrady/SpatialDengue: Fits and simulates a spatial model of the vector-borne disease dengue

Description

Runs multiple versions of DEN.spatial and calculates confidence intervals

Usage

DEN.spatial.ensemble(weekdates, fitdat, pastdat, unipix, pixdistmat,
  steprun, nruns, qtiles = c(0.025, 0.25, 0.5, 0.75, 0.975),
  ncores = 1, paramsList = NULL, seasonal = FALSE,
  seasonal_vector = NA)

Arguments

weekdates	Two element vector of the start and end weeks of the simulation over which the mdoel will be evaluated over
fitdat	Data frame of the locations, numbers and timings (in weeks) of cases to fit the model to, see ?sgdat
pastdat	Data frame of the locations, numbers and timings (in weeks) of all cases in the dataset (is used to generate the starting immunity profile), see ?sgdat
unipix	Universal pixel lookup table, see ?make.unipix
pixdistmat	A patch distance matrix, see example
steprun	integer, number of days for which the simulation should run, excluding burn in period
nruns	Integer, number of stochastic model runs over which to average over
qtiles	vector of quantiles over which results should be summarised (one or more values between 0 and 1)
ncores	number of cpu cores to run simulations over (in parallel). Uses snowfall for parallelisation. Specify cores > 1 (default = 1) for parallel processing
paramsList	Optional parameter list. If not supplied returns to defaults, see tutorial for full parameter list, see ?model.run for full list and explanation of parameters
seasonal	Logical, whether to use the defaul DEN.spatial function or the seasonal DEN.spatial.seasonal version. If this equals TRUE please supply a seasonal vector see ?DEN.spatial.seasonal
seasonal_vector	A vector of seasonal transmission intensity. See ?DEN.spatial.seasonal

Details

Returns a list with each component summarising the results across multiple model runs at a given quantile, e.g. 0.95, 0.5, 0.05, etc.

Examples

data(sgdat)
data(sgpop)
sgpop <- pop.process(sgpop, agg = 10)
unipix <- make.unipix(sgpop)
pixdistmat <- distm(cbind(unipix$x, unipix$y))
sgdat <- data.frame(sgdat, patchID = apply(cbind(sgdat[, 3:2]), 1, pix.id.find, unipix))
weekdates <- c(40, 92)
# run model ensemble with default parameters
denmod_sim <- DEN.spatial.ensemble(weekdates, sgdat, sgdat, unipix, pixdistmat, 365, 10)

obrady/SpatialDengue documentation built on Nov. 27, 2020, 12:13 p.m.