Skip to content

byRank()

Measure the Zipf slope of a set of counts, ranked from largest to smallest.

byRank(counts)

Ranks the counts from largest to smallest, then fits a trendline to the rank-against- count plot on a log-log scale. A slope close to -1 with a good fit indicates a Zipf-like distribution.

Parameters

byRank(counts)
Parameter Type Default Description
counts Iterable[int or float] required How often each item occurs.

Returns

return slope, r2, yint

Value Type Description
slope float The slope of the fitted trendline.
r2 float How well the trendline fits the data, from 0.0 to 1.0.
yint float The trendline's y-intercept.