R/tam_pv_mcmc_parameter_summary.R

In TAM: Test Analysis Modules

## File Name: tam_pv_mcmc_parameter_summary.R
## File Version: 0.14

tam_pv_mcmc_parameter_summary <- function(parameter_samples, level)
{
    require_namespace_msg("coda")
    NP <- ncol(parameter_samples)
    parnames <- colnames(parameter_samples)
    dfr <- data.frame("parm"=parnames,
                "est"=colMeans(parameter_samples) )
    dfr$se <- apply( parameter_samples, 2, stats::sd )
    #--- credibility interval
    dfr$low <- apply( parameter_samples, 2, stats::quantile, prob=(1-level)/2, na.rm=TRUE)
    dfr$upp <- apply( parameter_samples, 2, stats::quantile, prob=1 - (1-level)/2, na.rm=TRUE)
    some_miss <- sum( is.na(parameter_samples) ) > 0
    #--- effective sample size
    if ( some_miss){
        dfr$ESS <- NA
    } else {
        dfr$ESS <- round( coda::effectiveSize( x=parameter_samples ), 1)
    }
    #--- Rhat statistic
    dfr$Rhat <- tam_Rhat_3splits(parameter_samples=parameter_samples)
    #--- OUTPUT
    rownames(dfr) <- NULL
    return(dfr)
}