solarforecastarbiter.metrics.probabilistic.quantile_skill_score

solarforecastarbiter.metrics.probabilistic.quantile_skill_score(obs, fx, fx_prob, ref, ref_prob)[source]

Quantile Skill Score (QSS).

\[\text{QSS} = 1 - \text{QS}_{\text{fx}} / \text{QS}_{\text{ref}}\]

where \(\text{QS}_{\text{fx}}\) is the Quantile Score of the evaluated forecast and \(\text{QS}_{\text{ref}}\) is the Quantile Score of a reference forecast. [1]

Parameters:
  • obs ((n,) array_like) – Observations (physical unit).
  • fx ((n,) array_like) – Forecasts (physical units) of the right-hand-side of a CDF interval, e.g., fx = 10 MW is interpreted as forecasting <= 10 MW.
  • fx_prob ((n,) array_like) – Probability [%] associated with the forecasts.
  • ref ((n,) array_like) – Reference forecast (physical units) of the right-hand-side of a CDF interval.
  • ref_prob ((n,) array_like) – Probability [%] associated with the reference forecast.
Returns:

skill (float) – The Quantile Skill Score [unitless].

References

[1]Bouallegue, Pinson and Friederichs (2015) “Quantile forecast discrimination ability and value”, Quarterly Journal of the Royal Meteorological Society 141, pp. 3415-3424. doi: 10.1002/qj.2624

Notes

This function returns 0 if QS_fx and QS_ref are both 0.