risk.mean_excess()

The empirical mean-excess function e(u) = E[X - u | X > u] at each

Usage

risk.mean_excess(
    draws,
    thresholds,
)

threshold, linear above a threshold where a GPD fits.

Parameters

draws: list of float

thresholds: list of float

Returns

list of (float, float, int)

(u, e(u), number of draws above u); e(u) is NaN when none are.

Examples

>>> from prospicio.risk import mean_excess
>>> mean_excess([1.0, 2.0, 3.0, 4.0], [2.0])

[(2.0, 1.5, 2)]