Atmospheric chemistry and physics · 2007 · 45 citations · 33 references
BusinessDecision SciencesStatistical ModelingStochastic ModelingUncertainty QuantificationAir Pollution DispersionAtmospheric DispersionBayesian ModelingEnsemble MethodsProbabilistic ModelingStatistical InferenceAccidental Release ModelingBayes' TheoremDispersion ModelEnsemble AveragingLong-range TransportMulti-model EnsemblesPollutant Dispersion ModelingSystematic Uncertainties
Abstract. In this paper we present an approach for the statistical analysis of multi-model ensemble results. The models considered here are operational long-range transport and dispersion models, also used for the real-time simulation of pollutant dispersion or the accidental release of radioactive nuclides. We first introduce the theoretical basis (with its roots sinking into the Bayes theorem) and then apply this approach to the analysis of model results obtained during the ETEX-1 exercise. We recover some interesting results, supporting the heuristic approach called "median model", originally introduced in Galmarini et al. (2004a, b). This approach also provides a way to systematically reduce (and quantify) model uncertainties, thus supporting the decision-making process and/or regulatory-purpose activities in a very effective manner.
33
Independent component analysis: algorithms and applications
Aapo Hyvärinen, Erkki Oja · Neural Networks · 2000 · 8.8K citations
Using Bayesian Model Averaging to Calibrate Forecast Ensembles
Adrian E. Raftery, Tilmann Gneiting, Fadoua Balabdaoui et al. · Monthly Weather Review · 2005 · 2K citations · Full text
Bayesian Model Averaging for Linear Regression Models
1997 · 1.8K citations