distribution_check: Use simulations to compare the observed distribution with the...

In inbo/inlatools: Diagnostic Tools for INLA Models

Use simulations to compare the observed distribution with the modelled distribution

Description

Use simulations to compare the observed distribution with the modelled distribution

Usage

distribution_check(object, nsim = 1000, seed = 0L)

## S4 method for signature 'inla'
distribution_check(object, nsim = 1000, seed = 0L)

Arguments

object	the INLA model
nsim	the number of simulation
seed	See the same argument in ?inla.qsample for further information. In order to produce reproducible results, you ALSO need to make sure the RNG in R is in the same state, see example below. When seed is non-zero, num.threads is forced to "1:1" and parallel.configs is set to FALSE, since parallel sampling would not produce a reproducible sequence of pseudo-random numbers.

See Also

Other checks: dispersion_check(), fast_aggregation_check(), fast_distribution_check(), get_anomaly()

Examples

library(INLA)
model <- inla(
  poisson ~ 1,
  family = "poisson",
  data = data.frame(
    poisson = rpois(20, lambda = 10),
    base = 1
  ),
  control.predictor = list(compute = TRUE),
  control.compute = list(config = TRUE)
)
distribution_check(model, seed = 20181202)

inbo/inlatools documentation built on Oct. 10, 2024, 8:11 a.m.