R/formatZ.R

In miWQS: Multiple Imputation Using Weighted Quantile Sum Regression

#  In makeJournalTables: Don't export.
#   #   #'Formatting Frequencies and Percentages
#'
#' @family format
#' @keywords format
#'
#' @description Formats frequencies and percentages as Freq(Percent\%) like  6.00 (18.02\%).
#'
#' @details The columns of x can come from \code{\link[base]{table}} and
#'           \code{\link[base]{prop.table}}. The additional arguments format the
#'           frequencies and percentages.
#' @param n The frequency vector
#' @param ColPct The percentage vector
#' @inheritParams base::format
#' @param ... Additional formatting parameters passed to \code{\link[base]{format}}.
#' @return    A matrix consisting of formatted frequency/percentage
#'
#' @examples
#' # Example 1: A Toy Data Frame
#' formatZ (c(45, 17, 14), ColPct = c (78.9, 63.0, 46.7), digits = 3, nsmall = 1)
#'
#' # Example 2: Using table and prop.table functions.
#' data(CO2)
#' t <- table(CO2$Type, CO2$Treatment)
#' t
#' colPct <- 100 * prop.table(t, 2)
#' # Need to convert t into a vector to apply function; can use as.data.frame(), like
#' as.data.frame(t)
#' cbind(as.data.frame(t)[, 1:2], formatZ (as.data.frame(t)[, 3], as.data.frame(colPct)[, 3]))
#' # which can be easily copied and pasted into any professional table, like to a colleague or a
#' # peer-reviewed journal.
#'
#'
#'
#' #  In makeJournalTables: Don't export.
#' @noRd

formatZ <- function(n, ColPct, digits = 1, nsmall = 0, ...) {
  # Check the function
  stopifnot(length(n) == length(ColPct))

  # Format...
  pct <- paste0("(", format(ColPct, trim = TRUE, digits = digits, nsmall = nsmall, ...), "%)", ",")
  A <- matrix(paste (n, pct, sep = " "), ncol = 1)

  return(A)
}

# Archive
# Used to combine n and colPct into a matrix, x. But removed it from the function.
##   #'@param x  A data frame or a matrix with 2 columns or a list with 2 entries in the following order:
#   #'             a frequency vector, n, and a column percentage vector, ColPct. A warning is produced to ensure
#   #'             this is followed. Names are not necessary, but helpful.