R/as_rv_bugs.R

In jsta/rv: Simulation-Based Random Variable Objects

#' @export
#' @method as.rvsummary bugs
as.rvsummary.bugs <- function (x, list.=TRUE, ...) {
  if (list.) {
    lapply(as.rv(x, list.=TRUE), as.rvsummary, ...)
  } else {
    as.rvsummary(as.rv(x, list.=FALSE), ...)
  }
}



#' Coerce a bugs object into Random Variable Objects
#' 
#' \code{as.rv.bugs} coerces an \code{R2WinBUGS} object to a list of \code{rv}
#' objects or to a named rv object (vector).
#' 
#' \code{as.rvsummary.bugs} works similarly but coerces the resulting \code{rv}
#' objects into \code{rvsummary} objects.
#' 
#' 
#' @aliases as.rv.bugs as.rvsummary.bugs
#' @param x a bugs (R2WinBUGS) object
#' @param list. logical; return a list of \code{rv} objects instead of a single
#' \code{rv} object (vector)?
#' @param \dots (ignored)
#' @return If \code{list.=TRUE}, a named \emph{list} of random vectors or a
#' named random vector, otherwise a random vector.  (Usually one would prefer a
#' list.)f
#' @author Jouni Kerman \email{jouni@@kerman.com}
#' @references Kerman, J. and Gelman, A. (2007). Manipulating and Summarizing
#' Posterior Simulations Using Random Variable Objects. Statistics and
#' Computing 17:3, 235-244.
#' 
#' See also \code{vignette("rv")}.
#' @keywords classes
#' 
#' @export
#' @method as.rv bugs
as.rv.bugs <- function (x, list.=TRUE, ...) {
  # as.rv.bugs - transform a R2WinBUGS object into a random variable object
  # n.chains,
  # n.iter
  # n.burnin
  # n.thin
  # n.keep
  # n.sims
  # sims.array=sims.array[,,rank.long]
  # sims.list
  # sims.matrix=sims[,rank.long]
  # summary
  # mean=summary.mean
  # sd=summary.sd
  # median=summary.median
  # root.short
  # long.short
  # dimension.short
  # indexes.short
  # last.values
  #     sims <- c(bugs.sims(parameters.to.save, n.chains, n.iter, 
  #          n.burnin, n.thin, DIC), model.file = model.file, 
  #          is.DIC = DIC)
  #      class(sims) <- "bugs"
  #      return(sims)
  if (is.null(x['sims.array']))
    stop("Argument does not contain a 'sims.array'")
  A <- x$sims.array
  dnA <- dimnames(A)
  dim(A) <- c(prod(dim(A)[1:2]), dim(A)[3])
  dimnames(A) <- list(dnA[[1]], dnA[[3]])
  r <- rvsims(A)
  rvattr(r, 'Rhat')  <- x$summary[,'Rhat']
  rvattr(r, 'n.eff') <- x$summary[,'n.eff']
  if (list.) {
    return(splitbyname(r))
  } else {
    return(r)
  }
}