Assessing probabilistic forecasts of multivariate quantities, with an application to ensemble predictions of surface winds

By Tilmann Gneiting, Larissa I. Stanberry, E. Grimit, Leonhard Held, Nicholas Alexander Johnson. 2008

309 citations recorded by the source corpus.

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

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

Ensemble Model Output Statistics for Wind Vectors (2012, methodology): “Finally, we consider an error dressing ensemble as proposed by Gneiting et al. (2008) in the spirit of the work of Roulston and Smith (2003), where we dress the UWME mean forecast with 35 error vectors from the corresponding training set. d. Results over the Pacific Northwest We now give…”
Assessing the Calibration of High-Dimensional Ensemble Forecasts Using Rank Histograms (2013, methodology): “Following the general setup of Gneiting et al. (2008), the rank of the observation in S is calculated in two steps, (i) apply a prerank function ρS : Rd → R+ to calculate the prerank, ρS(x), of every x ∈ S; (ii) set the rank of the observation xm equal to the rank of ρS(xm) in {ρS(x1), . . . ,…”
Short‐Term Wind Speed Forecasting for Power System Operations (2012, methodology): “Gneiting et al. (2008) introduced methods to assess probabilistic forecasts of multivariate quantities: Box’s density ordinate transform works for density forecasts and the multivariate energy score can be used in comparing density forecasts using a proper loss function.”
Wind Energy: Forecasting Challenges for Its Operational Management (2013, background): “Going from univariate to multivariate aspects, Gneiting et al. (2008) explained how the previously introduced paradigm can be readily generalized for multivariate probabilistic predictions, yielding an evaluation framework including skill scores and diagnostic tools.”
Combining probability forecasts (2010, background): “Sharpness describes how far away the forecasts are from the naive, climatological baseline forecast, i.e. the marginal event frequency (Gneiting et al., 2008; Winkler and Jose, 2008).”
Discrimination ability of the Energy score (2013, methodology): “For the calculation of these Energy score values, 1000 random draws from both densities are used, then employing the computationally efficient estimator proposed by [13].”
Geostatistical Modelling Using Non‐Gaussian Matérn Fields (2015, methodology): “We displays four statistics from the cross-validation predictions: the variance of the residuals; the variance of the standardized residuals; the continuous ranked probability score (CRPS) (Matheson & Winkler, 1976), which probably is the most employed scoring role in probabilistic forecasts; and th”
Evaluating Probabilistic Forecasts with scoringRules (2017, methodology): “The CDFs for the two-piece normal distribution are given by Gneiting and Thorarinsdottir (2010) give an explicit CRPS formula.”
Validating Bayesian Inference Algorithms with Simulation-Based Calibration (2018, methodology): “An interesting extension of this methodology would be to incorporate some of the advances in multivariate calibration of probabilistic forecasts (Gneiting et al., 2008; Thorarinsdottir, Scheuerer and Heinz, 2013).”
Making and Evaluating Point Forecasts (2009, background): “Applications include those of Gneiting et al. (2008) and Hering and Genton (2010) to predictions of wind vectors, or that of Laurent, Rombouts, and Violante (2009) to forecasts of multivariate volatility, to name but a few.”
Local Proper Scoring Rules (2009, methodology): “For other examples of proper scoring rules on R d , we refer to Matheson and Winkler (1976), Dawid and Sebastiani (1998), Gneiting and Raftery (2007) and Gneiting et al. (2008).”