SECTION 15.23 ROBUST RANK-ORDER TEST

The robust rank-order test tests the hypothesis that the medians of two distributions are the same without assuming that the underlying distributions themselves are the same. See S. Siegel and N. J. Castellan, Nonparametric Statistics, 2nd ed., 1988, pp. 206-215. The call is



        CALL RROT(X,Y,M,N,U,IER)
where X = a REAL*8 array of sample values, dimensioned X(M) (input)
Y = a REAL*8 array of sample values, dimensioned Y(N) (input)
M = number of observations in sample 1 (input)
N = number of observations in sample 2 (input)
U = the robust rank order statistic (output)
IER = error return (=-1 array allocation error)
Return to

Beginning