Palma ratio: the top-10% share over the bottom-40% share.
The modern measure of extreme inequality, more sensitive than the Gini to the
tails: the total quantity held by the best-served 10% of stations divided by
that of the least-served 40%.
See Also
theil_index : A decomposable inequality alternative.
concentration_metrics : Gini + Theil + top-decile share in one call.
Examples:
>>> round(palma_ratio(list(range(1, 11))), 1)
1.0
Source code in src/gbfs_toolkit/analytics/distribution.py
| def palma_ratio(values) -> float:
"""Palma ratio: the top-10% share over the bottom-40% share.
The modern measure of *extreme* inequality, more sensitive than the Gini to the
tails: the total quantity held by the best-served 10% of stations divided by
that of the least-served 40%.
See Also
--------
[`theil_index`][gbfs_toolkit.theil_index] : A decomposable inequality alternative.
[`concentration_metrics`][gbfs_toolkit.concentration_metrics] : Gini + Theil + top-decile share in one call.
Examples
--------
>>> round(palma_ratio(list(range(1, 11))), 1)
1.0
"""
x = np.sort(np.asarray(values, dtype="float64"))
x = x[np.isfinite(x)]
n = x.size
if n == 0 or x.sum() == 0:
return float("nan")
bottom40 = x[: int(np.floor(0.4 * n))].sum()
top10 = x[int(np.ceil(0.9 * n)) :].sum()
if bottom40 <= 0:
return float("inf")
return float(top10 / bottom40)
|