Type I error of different resampling methods in multivariate count simulations from a lognormal-Poisson model, varying <i>n</i> (left) and <i>p</i> (right).

<p>Mean abundances were manipulated by multiplying the matrix of the means <b><i>μ</i></b> by a factor (a) <i>δ</i> = 1, (b) <i>δ</i> = 2.5, (c) <i>δ</i> = 5. The shaded grey region represents a 95% confidence band around the nominal Type I error rate of 0.05. The PIT-trap performed reasonably well in simulations varying <i>n</i>, but became inflated at large <i>p</i> because of failure of model assumptions, for low-medium abundances (a-b). The parametric bootstrap, assuming an unstructured correlation matrix, performed similarly.