论文部分内容阅读
Assuming a multivariate normal distribution, a distance test statistic is proposed to test the hypothesis of equivalence of normal means, where the equivalence is defined to be maximum distance of treatment means from the grand mean being falling into a small but negligible zone.The least favorable mean configuration (LFMC) to guarantee the maximum level at a null hypothesis is investigated.These level (and the power) of the test are fully independent of the unknown means and variances.For a given null hypothesis, the p-value of the test can be determined.Statistical tables of critical values and computer program for evaluating the p-value are provided.