5.2 Let T1 and T2 be two competing estimators of

and L(·, ·) be a lossfunction. Denote

Table 5.4 Powers of two-sample Hotelling T2 test (H), two-sample Wilcoxon test (W) and two-sample N−distance type test (N) based on distances for various m = n and = 0.05. The first sample X always has the two-dimensional Cauchy distribution.
The second sample Y is obtained as Y = μ + Y , where Y is generated as a two-dimensional Cauchy distribution independent on X. Values of m, n,μ and
are specified in the first column.