Skip to main content
Fig. 26 | Journal of Big Data

Fig. 26

From: Cumulative deviation of a subpopulation from the full population

Fig. 26

Stanislaus County, reporting the number of related children in the household, with scores being \(\log _{10}\) of the adjusted household income; \(n =\) 1624; Kuiper’s statistic is \(0.1489 / \sigma = 4.500\), Kolmogorov’s and Smirnov’s is \(0.1467 / \sigma = 4.433\). The reliability diagrams are all commensurate with each other, but that is clear only after careful comparison of them all. Making sense of the reliability diagrams in this case would seem to require generating multiple plots, with several different numbers of bins. The scalar summary statistics have little trouble detecting statistically significant deviation here

Back to article page