An asymptotically optimal nonpammetrf c statistic foe testing equality of two normal population mea1s and ¥a1xance5

A'nonparametric two-sample test is introduced which Is asymp¬totically optimal for testing the hypothesis that two samples come from identical normal populations versus the alternative that the populations are normal, but vary in the mean and/or variance. The statistic is shown to be robust when the populations arie not nor¬mal, Finite sample properties are considered and the exact criti¬cal points obtained for certain sample sizes