Group intervals for Rao-Robson-Nikulin´s criterion selection with help of noncentrality parameter maximization

The paper is devoted to the investigation of optimal grouping for Rao-Robson-Nikulin´s criteria. The points were found for optimal grouping via noncentrality parameter of statistics distribution maximization under the alternative hypothesis. The comparison with other methods was conducted with the help of computer simulation for two pairs of hypotheses: the first was the case of normal and logistic distribution and the second was the case of two normal distributions with different sample size. The scale parameter was chosen for estimation. Conclusions about criteria statistics distribution under the alternative hypothesis were made by considering simulation results.