2 DATAPOINTS and A MODEL:
(x=1,y=2) and (x=2,y=8) are 2 indep. obsns. from no-intercept model E[y|x] = beta x.
y might be the total numbers of errors on x randomly sampled pages of a document,
e.g., 2 errors in x=1 page, and 8 errors in total in a separate
sample of x = 2 pages.
Beta is the mean no. of errors per page of
the document. .... Or, y might be the total weight of x randomly
sampled pages of a document: 2 units of weight in total for
a sample of x = 1 page, and 8 units for a separate sample of x = 2 pages,
Beta is the mean weight per page.
Why did statisticians, epidemiologists,
and grade 6 students gave different estimates, e.g., 3.6, 3.33 and 3.46/page?
.