In this paper we develop a novel bootstrap test for the comparison of two multinomial distributions. The two distributions are called {\it equivalent} or {\it similar} if a norm of the difference between the class probabilities is smaller than a given threshold. In contrast to most of the literature our approach does not require differentiability of the norm and is in particular applicable for the maximum- and $L^1$-norm.

翻译：本文提出了一种新颖的自助检验方法，用于比较两个多项分布。若两个分布类别概率之差的范数小于给定阈值，则称其具有等价性或相似性。与大多数文献不同，本方法不要求范数具有可微性，尤其适用于最大范数和$L^1$范数情形。