Calculate Bayes factor from log marginal likelihoods

Calculate Bayes factor from log marginal likelihoods

Usage

bayes_factor(logml1, logml2)

Arguments

logml1

Log marginal likelihood of Model 1 (e.g., target model)

logml2

Log marginal likelihood of Model 2 (e.g., reference/null model)

Value

An object containing Bayes factor, log Bayes factor, estimation error, and interpretation

Examples

if (FALSE) { # \dontrun{
  # Compare two models using Bayes Factor
  fit1 <- rtmb_lm(mpg ~ wt, data = mtcars)
  fit2 <- rtmb_lm(mpg ~ wt + hp, data = mtcars)
  mcmc1 <- fit1$sample(sampling = 500, warmup = 500)
  mcmc2 <- fit2$sample(sampling = 500, warmup = 500)
  bf <- bayes_factor(mcmc1, mcmc2)
  print(bf)
} # }