Decomposition of the Continuous Ranked Probability Score for Ensemble Prediction Systems

By Hans Hersbach. 2000

1,454 citations recorded by the source corpus.

Subfields: Atmospheric Science, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/4342b6334a1e67775278c12040a24cba5b05b473

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

In Global and Planetary Change

Citations in the corpus

Recent advances in electricity price forecasting: A review of probabilistic forecasting (2016, background): “The CRPS has several appealing properties [107]: (i) its definition does not require the”
Estimating reliability and resolution of probability forecasts through decomposition of the empirical score (2012, methodology): “For mathematical proofs, the interested reader is deferred to the classical paper by Murphy (1973) for the Brier score (and for binary forecasting problems), to Hersbach (2000) for the Continuous Ranked Probability Score (not discussed here), and to Bröcker (2009) for the general case.”
Distributional regression forests for probabilistic precipitation forecasting in complex terrain (2018, methodology): “Hersbach, H. (2000). Decomposition of the Continuous Ranked Probability Score for Ensemble Prediction Systems. Weather and Forecasting 15 559–570. Hofner, B., Mayr, A. and Schmid, M. (2016). gamboostLSS: An R Package for Model Building and Variable Selection in the GAMLSS Framework. Journal of Stati”
The forecast skill horizon (2015, methodology): “The CRPS has been computed as in Hersbach (2000).”
Probabilistic quantitative precipitation forecasting using Ensemble Model Output Statistics (2013, methodology): “Both of these aspects make a strong case for the continuous ranked probability score (2) (Hersbach, 2000; Gneiting and Raftery, 2007).”
Interpretation of Rank Histograms for Verifying Ensemble Forecasts (2001, methodology): “The rank histogram, when correctly used and interpreted, measures the reliability of the ensemble (Talagrand et al. 1997; Hersbach 2000).”
Probabilistic Forecasting with Spline Quantile Function RNNs (2019, background): “Other equivalent deﬁni-tions of the CRPS exist, e.g. as the average of Brier scores, | � | � | � | (Hersbach, 2000; Gneiting and Raftery, 2007) for more details. where 2 R is the intercept term, 2 R are weights describing the slopes of the function pieces, 2 R is a vector of positions, i.e., the…”
Forecast verification for extreme value distributions with an application to probabilistic peak wind prediction (2012, background): “The continuous ranked probability score (CRPS) assesses the predictive skill of a forecast in terms of the entire predictive distribution F (Unger, 1985; Hersbach, 2000) and can thus assess both the calibration and the sharpness of the forecast simultaneously.”
The continuous ranked probability score for circular variables and its application to mesoscale forecast ensemble verification (2006, background): “…forecast uncertainty is not yet well predicted at mesoscale resolutions near the surface, despite the inclusion of multi-scheme physics diversity and surface boundary parameter perturbations in the mesoscale ensemble design.1 Keywords: Proper scoring rule; von Mises distribution; Wind direction.”
Skillful Twelve Hour Precipitation Forecasts using Large Context Neural Networks (2021, methodology): “The high CRPSS for instantaneous rate that MetNet-2 obtains relative to HRRR, from 0.65 at one hour of lead time to 0.35 at 12 hours of lead time, indicates the ability of MetNet-2 to estimate forecast probabilities well when evaluated against the ground truth.”
Evaluating raw ensembles with the continuous ranked probability score (2012, result): “This would then contradict the results of Hersbach (2000) which imply that, of two forecasts with the same information content, an unreliable forecast cannot have a better score than a reliable one.”
The Multiensemble Approach: The NAEFS Example (2009, methodology): “The continuous ranked probabilistic score (CRPS; Stanski et al. 1989) measures the global skill of an EPS, and its reliabilityresolution partition (Hersbach 2000) enables the validation of these two attributes of a probabilistic system (Toth et al. 2003).”
Diffusion Probabilistic Modeling for Video Generation (2022, background): “A brief introduction about the metric is available in Appendix A These metrics are frequently used in probabilistic forecasting problems in, e.g., meteorology or ﬁnance [87,88].”
All age–depth models are wrong, but are getting better (2015, methodology): “We also used the continuous ranked probability score (CRPS; Gneiting and Raftery, 2007; Hersbach, 2000) to assess the quality of age–depth models.”
Empirical Evaluation of Hierarchical Ground‐Motion Models: Score Uncertainty and Model Weighting (2017, background): “Nevertheless, nonlocal proper scores (e.g., Hersbach, 2000) have also been widely used in forecast evaluation.”