R/pmle_process_prior.R

In alexanderrobitzsch/LAM: Some Latent Variable Models

## File Name: pmle_process_prior.R
## File Version: 0.23

###########################################################
pmle_process_prior <- function( prior, pars ){
    #*** convert string to prior list
    if ( is.character(prior) ){
        prior <- sirt::prior_model_parse( prior_model=prior )
    }
    #*** correct list specification
    parnames <- names(pars)
    NP <- length(parnames)
    prior1 <- as.list(1:NP)
    names(prior1) <- parnames

    #**** correct prior specification
    prior0 <- list( "dimproper", list(NA) )
    parnames0 <- setdiff( parnames, names(prior) )
    NP0 <- length(parnames0)
    if (NP0 > 0 ){
        for (pp in 1:NP0){
            prior[[ parnames0[pp] ]] <- prior0
        }
    }
    #**** write prior
    for (pp in 1:NP){
        prior1[[pp]] <- prior[[ parnames[pp] ]]
    }
    dens1 <- pmle_prior_extract_density( prior=prior1 )
    dens1 <- data.frame("parameter"=names(pars), "prior"=dens1 )
    rownames(dens1) <- NULL
    res <- list("prior"=prior1, "dens"=dens1 )
    return(res)
}
###########################################################