Skip to main content
Fig. 25 | Journal of Big Data

Fig. 25

From: Cumulative deviation of a subpopulation from the full population

Fig. 25

Napa County, reporting whether the household has access via a satellite dish to the internet, with scores being \(\log _{10}\) of the adjusted household income; \(n =\) 679; Kuiper’s statistic is \(0.02761 / \sigma = 2.259\), Kolmogorov’s and Smirnov’s is \(0.02695 / \sigma = 2.205\). The intense deviation around scores of 4.6 is apparent in the reliability diagrams with 10 and 20 bins each, but the latter resolves the spike much better while being unfortunately too noisy for many other scores. The plot of cumulative differences resolves the sharp jump around scores of 4.6 without detracting from the display at other scores. The scalar summary statistics report only very mildly statistically significant deviation, unable to fully account for the high deviation around scores of 4.6

Back to article page