Two-Sample Permutation Test for Comparing Marginal Probabilities of Multivariate Bernoulli Distributions

The contribution of this paper is both methodological and applicative. A combined permutation test for two independent samples is proposed for comparing the vectors of marginal probabilities of two multivariate Bernoulli distributions. The power behavior of the test is analyzed through a Monte Carlo simulation study. The application of the method to real data concerning the propensity towards circular economy of small Italian companies in the sector of metal, contributes to the debates on the effect of firm size to the “circularity” of companies.