R/cor-mtest.R

In corrplot: Visualization of a Correlation Matrix

#' Significance test which produces p-values and confidence intervals for each
#' pair of input features.
#'
#' @param mat Input matrix of size \code{NxF},
#'   with \code{N} rows that represent samples
#'   and \code{F} columns that represent features.
#' @param \dots Additional arguments passed to function \code{\link{cor.test}},
#'   e.g. \code{conf.level = 0.95}.
#'
#' @return Return a list containing:
#'   \item{p}{Square matrix of size \code{FxF} with p-values as cells}
#'   \item{lowCI}{Square matrix of size \code{FxF}, each cell represents the
#'   \emph{lower part} of a confidence interval}
#'   \item{uppCI}{Square matrix of size \code{FxF}, each cell represents the
#'   \emph{upper part} of a confidence interval}
#'
#' @seealso Function \code{\link{cor.test}}
#'
#' @keywords p-value confidence significance
#' @export
cor.mtest = function(mat, ...) {
  mat = as.matrix(mat)
  n = ncol(mat)
  p.mat = lowCI.mat = uppCI.mat = matrix(NA, n, n)
  diag(p.mat) = 0
  diag(lowCI.mat) = diag(uppCI.mat) = 1
  for (i in 1:(n - 1)) {
    for (j in (i + 1):n) {

      tmp = cor.test(x = mat[, i], y = mat[, j], ...)
      p.mat[i, j] = p.mat[j, i] = tmp$p.value

      # only 'pearson' method provides confidence intervals
      if (!is.null(tmp$conf.int)) {
        lowCI.mat[i, j] = lowCI.mat[j, i] = tmp$conf.int[1]
        uppCI.mat[i, j] = uppCI.mat[j, i] = tmp$conf.int[2]
      }
    }
  }

  colnames(p.mat) = rownames(p.mat) = colnames(mat)
  colnames(lowCI.mat) = rownames(lowCI.mat) = colnames(mat)
  colnames(uppCI.mat) = rownames(uppCI.mat) = colnames(mat)

  list(
    p = p.mat,
    lowCI = lowCI.mat,
    uppCI = uppCI.mat
  )
}