R/d_stat.R

In qiyuan98/energy.test: three kinds of hypothesis tests based on energy statistics

#' calculate the energy statistics from balanced designed samples
#'
#' This fucntion calculate the statistics to test multrivariate test for equal distributions based on energy method
#' @param size the sample size
#' @param group the number of the groups of the samples
#' @param index for permutation test, indicate how to divide the samples into groups
#' @param distance the distance matrix
#' @return energy statistics used for multivariate test for equal distribution
d_stat <- function(size,group,index,distance){
  total <- size * group
  ind <- index;ind <- matrix(ind,byrow = TRUE,nrow = group)
  add.w <- numeric(group)
  for (i in 1:group) {
    add.w[i] <- size/2 * mean(distance[ind[i,],ind[i,]])
  }
  w <- sum(add.w)
  s <- numeric(1)
  for(i in 1:(group-1)){
    for(j in (i+1):group){
      s = s + size^2/total *
        ( 2*mean(distance[ind[i,],ind[j,]]) -
            mean(distance[ind[j,],ind[j,]]) -
            mean(distance[ind[i,],ind[i,]]) )
    }
  }
  s/(group - 1)/(w/(total - group))
}