Calibrated Probabilistic Forecasting Using Ensemble Model Output Statistics and Minimum CRPS Estimation

By Tilmann Gneiting, Adrian E. Raftery, Anton H. Westveld, T. Goldman. 2005

1,087 citations recorded by the source corpus.

Subfields: Atmospheric Science, Environmental Engineering, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/22cfa3fa5bc80e51ade7ff3889571e472a230cb1

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 Environmental Engineering

In Global and Planetary Change

Citations in the corpus

Evaluation of a nonparametric post‐processor for bias correction and uncertainty estimation of hydrologic predictions (2013, methodology): “27, 83–105 (2013) advantage of statistical post-processing and part of the overall gain in skill when using CBP-ICK (refer to Gneiting et al., 2005 also).”
Strictly Proper Scoring Rules, Prediction, and Estimation (2007, methodology): “Gneiting et al. (2005) estimated a predictive regression model using the optimum score estimator based on the CRPS—a choice motivated by the meteorological problem.”
Forecast verification for extreme value distributions with an application to probabilistic peak wind prediction (2012, methodology): “G/ follows a normal distribution, the CRPS of the lognormal GLM is calculated using the closed form expression in Gneiting et al. (2005).”
A Comparison between Raw Ensemble Output, (Modified) Bayesian Model Averaging, and Extended Logistic Regression Using ECMWF Ensemble Precipitation Reforecasts (2010, methodology): “…of statistical methods, such as logistic regression (LR; Brelsford and Jones 1967; Wilks 2006b), Bayesian model averaging (BMA; Raftery et al. 2005), nonho-mogeneous Gaussian regression (NGR; Gneiting et al. 2005), and Gaussian ensemble dressing (GED; Roulston and Smith 2003; Wang and Bishop 2005).”
Distributional regression forests for probabilistic precipitation forecasting in complex terrain (2018, methodology): “The three alternatives are: a stan- dard ensemble model output statistics approach (EMOS, Gneiting et al. 2005) based on an NGR, a GAMLSS with regressors prespecified based on meteorological expertise (following Stauffer, Umlauf, Messner, Mayr, and Zeileis 2017b), and a boosted GAMLSS (Hofner,…”
Wind power forecasting : state-of-the-art 2009. (2009, methodology): “[289] present a technique that is used to model the predicted distribution as a Gaussian, where”
Forecasting: theory and practice (2020, background): “, 1997), model output statistics (Glahn and Lowry, 1972; Wilkd, 2005), ensemble model output statistics (Gneiting et al., 2005), and model conditional processor (Todini, 2008; Coccia and Todini, 2011).”
Spatial Postprocessing of Ensemble Forecasts for Temperature Using Nonhomogeneous Gaussian Regression (2014, methodology): “The univariate ensemble BMAmethod of Raftery et al. (2005) is a kernel dressing approach where, for temperature, each bias-corrected ensemble member is dressed”
WeatherBench 2: A Benchmark for the Next Generation of Data‐Driven Global Weather Models (2023, methodology): “…producing an ensemble of generative roll‐outs, or they could directly predict probabilistic outcomes of interest, such as the probability distribution of precipitation as done in the MetNet family of models (Andrychowicz et al., 2023) or many post‐ processing models (e.g., Gneiting et al., 2005).”
Ensemble size: How suboptimal is less than infinity? (2018, methodology): “A transformation into a Gaussian or some other parametric distribution is performed, for instance, in ensemble model output statistics methods (EMOS: see, for example, Gneiting et al., 2005; Hemri et al., 2014).”
Probabilistic quantitative precipitation forecasting using Ensemble Model Output Statistics (2013, methodology): “To achieve this, we adapt the non-homogeneous Gaussian regression approach by Gneiting et al. (2005), so as to respect the peculiarities of precipitation.”
Ensemble forecasting (2008, background): “The reliability of an ensemble can be improved through calibration based on statistics of past ensemble forecasts for the same season and region [21,22].”
Energy Forecasting: A Review and Outlook (2020, methodology): “To that end, various post-processing techniques, such as ensemble model output statistics [70] or forecast combination [71], are used to calibrate and sharpen the initial forecasts.”
Comparison of Ensemble-MOS Methods Using GFS Reforecasts (2007, methodology): “Similarly, the more elaborate LR(2) and NGR models in Eqs.”