Fig. 11
From: Cumulative deviation of a subpopulation from the full population
Night snake (Hypsiglena torquata), with scores being the negative log-likelihoods; \(n =\) 1300; Kuiper’s statistic is \(0.1365 / \sigma = 11.35\), Kolmogorov’s and Smirnov’s is \(0.1365 / \sigma = 11.35\). Bins equispaced across either the subpopulation’s observations or the scores (where the scores are negative log-likelihoods in this figure) cannot resolve the severe deviations at low scores without being overly noisy elsewhere. The scalar summary statistics extremely successfully detect the statistically highly significant deviation of the subpopulation from the full population