GOODNESS-OF-FIT TEST ON THE BASIS OF THE CONFIDENCE INTERVALS OF THE EMPIRICAL PROBABILITY DISTRIBUTION FUNCTION
A goodness-of-fit test is constructed as a sequence of bilateral tests. It is shown that the central element of the sequence is the unbiased locally most powerful test. To calculate the probability distribution free significance level we offered to use an algorithm. The quality of the new goodness-of-fit test was evaluated with numerical modeling.
Keywords: goodness-of-fit test, interval estimate, probability distribution law, significance level
