“…The parameter settings we used are the same as in Wu et al [4]: the sample sizes were set as (n 1 , n 2 ) = (5, 5), (5, 10), (10,10), population means were set as ( 1 , 2 ) = (2, 1), and took values as (0.25, 0.5, 0.75, 1, 2, 5). For each parameter setting, 10 000 random samples were generated, and within each of the 10 000 random samples, 5000 R 's were obtained using the algorithm presented in Section 2.2.…”