Smooth Versions of the Mann–Whitney–Wilcoxon Statistics

The well-known Mann–Whitney–Wilcoxon (MWW) statistic is based on empirical distribution estimates. However, the data are often drawn from smooth populations. Therefore, smoothness characteristic not preserved. In addition, several authors have pointed out that an inadmissible estimate. Thus, in this work, we develop versions of MWW function This approach preserves characteristics and allows efficiency procedure to improve. our shown be robust against a large class dependent observations. Hence, by choosing rectangular array known functions, test lot more reflective real data.