rbi.helpers (version 0.4.0)

DIC: Compute Deviance Information Criterion (DIC) for a libbi model

Description

Computes the DIC of a libbi object containing Monte-Carlo samples. The effective number of parameters is calculated following Gelman et al., Bayesian Data Analysis: Second Edition, 2004, p. 182.

Usage

# S3 method for libbi
DIC(x, bootstrap = 0, ...)

Value

DIC

Arguments

x

a libbi object

bootstrap

number of bootstrap samples to take, 0 to just take data

...

any parameters to be passed to `bi_read` (e.g., `burn`)

Author

Sebastian Funk

Examples

example_run <- rbi::bi_read(
  system.file(package = "rbi", "example_output.nc")
)
example_model_file <- system.file(package = "rbi", "PZ.bi")
example_bi <- rbi::attach_data(
  rbi::libbi(example_model_file), "output", example_run
)
DIC(example_bi)