R/Dummy.R

In SciencesPo: A Tool Set for Analyzing Political Behavior Data

#' @encoding UTF-8
#' @title Generate Dummy Variables in a Data Frame
#'
#' @description Quickly create dummies in a data frame.
#'
#' @param x the column position to generate dummies
#' @param .data the a data.frame.
#' @param drop A logical value. If \code{TRUE}, unused levels will be omitted.
#'
#' @author Daniel Marcelino, \email{dmarcelino@@live.com}
#'
#' @details A matrix object
#'
#' @keywords Models
#'
#' @examples
#' df <- data.frame(y = rnorm(10), x = runif(10,0,1), sex = sample(1:2, 10, rep=TRUE))
#'
#' Dummify(df$sex) ;
#'
#' cbind(df, Dummify(df$sex));
#' @export
`Dummify` <-
  function (x, .data = NULL, drop = TRUE)
  {
    if (is.null(.data)) {
      varname <- as.character(sys.call(1))[2]
      varname <- sub("^(.*\\$)", "", varname)
      varname <- sub("\\[.*\\]$", "", varname)
    }
    else {
      if (length(x) > 1)
        stop("More than one variable to create dummies at same  time.")
      varname <- x
      x <- .data[, varname]
    }
    if (drop == FALSE && class(x) == "factor") {
      x <- factor(x, levels = levels(x), exclude = NULL)
    }
    else {
      x <- factor(x, exclude = NULL)
    }
    if (length(levels(x)) < 2) {
      warning(varname, " has only 1 dimension. Generating dummy variable anyway.")
      invisible(matrix(rep(1, length(x)), ncol = 1, dimnames = list(rownames(x), c(paste(varname, "_", x[[1]], sep = "")))))
    }
    mat <- model.matrix(~x - 1, stats::model.frame(~x - 1), contrasts = FALSE)
    colnames.mm <- colnames(mat)
    cat(" ", varname, ":", ncol(mat), "dummy variables generated\n")
    mat <- matrix(as.integer(mat), nrow = nrow(mat), ncol = ncol(mat), dimnames = list(NULL, colnames.mm))
    colnames(mat) <- sub("^x", paste(varname, "_", sep = ""), colnames(mat))
    if (!is.null(row.names(.data)))
      rownames(mat) <- rownames(.data)
    invisible(mat)
  }
NULL