Equivalence and noninferiority tests for quality, manufacturing and test engineers (Q2851665)

The book describes a way different from traditional hypotheses testing an approach how to set null hypotheses and alternatives. The test, based on the sample data, is used to decide whether we should believe that the parameter(s) lies (lie) outside the range of ``equivalence'' or ``noninferiority'' (null hypothesis) or whether it is better to believe that the parameter(s) is (are) within the range of equivalence or noninferiority (alternative hypothesis). The goal of the statistical test is to find an evidence that the parameter(s) is (are) ``close enough'' to ideal. The tests for basic problems that are encountered in statistics with a special attention to engineering problems, e.g., tests for paramater(s) of binomial distribution, tests for mean(s) and variance(s) of normal distribution, tests for parameter(s) of exponential distribution, tests for capability indices, tests for mean(s) of multivariate normal distribution, tests applied in reliability theory etc., are described and the corresponding critical regions are presented.