We propose a new proximity measure between samples which is based on confidence limits for the bulk of general population. The confidence limits are constructed by means of order statistics. For this proximity measure, we compute approximate confidence limits corresponding to a given significance level in the cases where the null hypothesis of the equality of hypothetical distribution functions may be true as well as false. We compare considered proximity measure with the Kolmogorov - Smirnov statistics and the Wilcoxon statistics for samples from various populations. On the basis of the proposed proximity measure, we construct statistical criterion for testing hypothesis of the equality of hypothetical distribution functions.
展开▼
