R/med1way.crit.R

In WRS2: A Collection of Robust Statistical Methods

med1way.crit <- function(n,alpha=.05,iter=1000,TEST=NA,SEED=TRUE){
#
#  Determine the critical value for the function
#  med1way, assuming normality, based on the sample
#  sizes in n.
#
  J<-length(n)
  x<-list()
  w<-vector("numeric",J)
  xbar<-vector("numeric",J)
  if(SEED)set.seed(2)
  chk<-NA
  grp<-c(1:J)
  for (it in 1:iter){
    for(j in 1:J){
      x[[j]]<-rnorm(n[j])
      w[j]<-1/msmedse(x[[grp[j]]])^2
      xbar[j]<-median(x[[grp[j]]])
      n[j]<-length(x[[grp[j]]])
    }
    u<-sum(w)
    xtil<-sum(w*xbar)/u
    chk[it]<-sum(w*(xbar-xtil)^2)/(J-1)
  }
  chk<-sort(chk)
  iv<-round((1-alpha)*iter)
  crit.val<-chk[iv]
  pval<-NA
  if(!is.na(TEST))pval<-sum((TEST<=chk))/iter
  list(crit.val=crit.val,alpha = alpha, p.value=pval)
}