Extract Parameters with Maximum Log Posterior

This function finds and extracts the parameter set corresponding to the maximum log posterior value from MCMC output. It assumes that the MCMC results object includes a coda_pars matrix or data frame with columns for log_prior, log_likelihood, and log_posterior, along with parameter names.

Usage

extract_max_posterior_params(results)

Arguments

results

A list containing MCMC output, with a coda_pars element (a matrix or data frame) that includes a column named "log_posterior" and columns for sampled parameters.

Value

A named numeric vector of parameter values corresponding to the maximum log posterior.

Examples

# Create dummy MCMC output for illustration
set.seed(123)
results <- list(
  coda_pars = data.frame(
    param1 = rnorm(100),
    param2 = rnorm(100),
    log_prior = rnorm(100),
    log_likelihood = rnorm(100),
    log_posterior = rnorm(100)
  )
)

# Extract parameters corresponding to the maximum log posterior
extract_max_posterior_params(results)
#>      param1    param2
#> 27 0.837787 0.2353866