runFieldModel: Run a Model and Estimate Underlying Field From Point and...

In nmmarquez/PointPolygon: Evaluate Point and Polygon Data Simulatneously

Description

Run a model and estimate underlying latent and probability field from point and polygon data using TMB.

Usage

runFieldModel(field, pointDF = NULL, polyDF = NULL, moption = 0,
  verbose = FALSE, symbolic = TRUE, control = list(eval.max = 10000,
  iter.max = 10000), rWidth = NULL, priors = FALSE, mcmc = FALSE,
  AprojPoly = NULL, shape3 = NULL, start = list(), ...)

Arguments

field	field object which simulated underlying data
pointDF	data simulated from samplePoints
polyDF	data simulated from samplePolygns
moption	int, intger indicating how polygon data should be estimated 0 is by mixed model approximation, 1 is by redistribution, 2 is by Utazi approach, 3 is by riemman approximation, 4 is ignoring polygon data, 5 is if all data was known.
verbose	logical, print model fitting information
symbolic	logical, use metas reordering in model fitting
control	list, control list passed to nlminb
rWidth	int, only used in moption 2 to build besag prior
priors	logical, default FALSE Should priors be evaluated for top level parameters.
mcmc	logical, default FALSE Should model be fit with MCMC. Not compatible with moption 2.
AprojPoly	sparseMatrix, sparse matrix with population weight information for polygons.
shape3	shape to build the adjaceny matrix for when moption == 2
start	starting points of parameters.
...	Further arguments to pass to tmbstan

Value

List of fitted model objects.

Examples

## Not run: 
unitSim <- simField(
N = 500, rangeE = .7,
offset = c(0.1, 0.2), 
max.edge = c(0.1,0.2),
beta0 = -2,
betaList = list(
    list(type="random", value=2),
    list(type="spatial", value=-.5),
    list(type="cluster", value=-2)
))

pointDF <- samplePoints(unitSim, 500, 100)

runFieldModel(unitSim, pointDF)


## End(Not run)

nmmarquez/PointPolygon documentation built on Dec. 10, 2020, 1:15 a.m.