rSPDE result extraction from INLA estimation results

Extract field and parameter values and distributions for an rspde effect from an inla result object.

Usage

rspde.result(
  inla,
  name,
  rspde,
  compute.summary = TRUE,
  parameterization = "detect",
  n_samples = 5000,
  n_density = 1024
)

Arguments

inla: An inla object obtained from a call to inla().
name: A character string with the name of the rSPDE effect in the inla formula.
rspde: The inla_rspde object used for the effect in the inla formula.
compute.summary: Should the summary be computed?
parameterization: If 'detect', the parameterization from the model will be used. Otherwise, the options are 'spde', 'matern' and 'matern2'.
n_samples: The number of samples to be used if parameterization is different from the one used to fit the model.
n_density: The number of equally spaced points to estimate the density.

Value

If the model was fitted with matern parameterization (the default), it returns a list containing:

marginals.range: Marginal densities for the range parameter
marginals.log.range: Marginal densities for log(range)
marginals.std.dev: Marginal densities for std. deviation
marginals.log.std.dev: Marginal densities for log(std. deviation)
marginals.values: Marginal densities for the field values
summary.log.range: Summary statistics for log(range)
summary.log.std.dev: Summary statistics for log(std. deviation)
summary.values: Summary statistics for the field values

If compute.summary is TRUE, then the list will also contain

summary.kappa: Summary statistics for kappa
summary.tau: Summary statistics for tau

If the model was fitted with the spde parameterization, it returns a list containing:

marginals.kappa: Marginal densities for kappa
marginals.log.kappa: Marginal densities for log(kappa)
marginals.log.tau: Marginal densities for log(tau)
marginals.tau: Marginal densities for tau
marginals.values: Marginal densities for the field values
summary.log.kappa: Summary statistics for log(kappa)
summary.log.tau: Summary statistics for log(tau)
summary.values: Summary statistics for the field values

If compute.summary is TRUE, then the list will also contain

summary.kappa: Summary statistics for kappa
summary.tau: Summary statistics for tau

For both cases, if nu was estimated, then the list will also contain

marginals.nu: Marginal densities for nu

If nu was estimated and a beta prior was used, then the list will also contain

marginals.logit.nu: Marginal densities for logit(nu)
summary.logit.nu: Marginal densities for logit(nu)

If nu was estimated and a truncated lognormal prior was used, then the list will also contain

marginals.log.nu: Marginal densities for log(nu)
summary.log.nu: Marginal densities for log(nu)

If nu was estimated and compute.summary is TRUE, then the list will also contain

summary.nu: Summary statistics for nu

Examples

# \donttest{
# devel version
if (requireNamespace("INLA", quietly = TRUE)) {
  library(INLA)

  set.seed(123)

  m <- 100
  loc_2d_mesh <- matrix(runif(m * 2), m, 2)
  mesh_2d <- inla.mesh.2d(
    loc = loc_2d_mesh,
    cutoff = 0.05,
    max.edge = c(0.1, 0.5)
  )
  sigma <- 1
  range <- 0.2
  nu <- 0.8
  kappa <- sqrt(8 * nu) / range
  op <- matern.operators(
    mesh = mesh_2d, nu = nu,
    range = range, sigma = sigma, m = 2,
    parameterization = "matern"
  )
  u <- simulate(op)
  A <- inla.spde.make.A(
    mesh = mesh_2d,
    loc = loc_2d_mesh
  )
  sigma.e <- 0.1
  y <- A %*% u + rnorm(m) * sigma.e
  Abar <- rspde.make.A(mesh = mesh_2d, loc = loc_2d_mesh)
  mesh.index <- rspde.make.index(name = "field", mesh = mesh_2d)
  st.dat <- inla.stack(
    data = list(y = as.vector(y)),
    A = Abar,
    effects = mesh.index
  )
  rspde_model <- rspde.matern(
    mesh = mesh_2d,
    nu.upper.bound = 2
  )
  f <- y ~ -1 + f(field, model = rspde_model)
  rspde_fit <- inla(f,
    data = inla.stack.data(st.dat),
    family = "gaussian",
    control.predictor =
      list(A = inla.stack.A(st.dat))
  )
  result <- rspde.result(rspde_fit, "field", rspde_model)
  summary(result)
}
#> Warning: the mean or mode of nu is very close to nu.upper.bound, please consider increasing nu.upper.bound, and refitting the model.
#>              mean        sd  0.025quant    0.5quant 0.975quant        mode
#> tau    0.00813859 0.0163503 1.22140e-11 6.80266e-04   0.059179 1.22140e-11
#> kappa 23.93130000 9.5804500 1.36645e+01 2.12293e+01  49.540100 1.63086e+01
#> nu     1.56109000 0.4196600 6.85689e-01 1.70615e+00   1.998480 1.99997e+00
# devel.tag
# }