KIT | KIT-Bibliothek | Impressum
Open Access Logo
DOI: 10.5445/IR/1000055628

Probabilistic forecasting and comparative model assessment, with focus on extreme events

Lerch, Sebastian

Abstract (englisch):
Probabilistic forecasts allow to quantify the prediction uncertainty and are essential for informed decision making. We investigate how to evaluate probabilistic forecasts with an emphasis on extreme events, and how to make and evaluate forecasts based on simulation output in Bayesian forecasting models. Further, we propose new models and estimation approaches for statistical post-processing of ensemble forecasts in numerical weather prediction.

Zugehörige Institution(en) am KIT Institut für Stochastik (STOCH)
Publikationstyp Hochschulschrift
Jahr 2016
Sprache Englisch
Identifikator URN: urn:nbn:de:swb:90-556285
KITopen ID: 1000055628
Verlag Karlsruhe
Abschlussart Dissertation
Fakultät Fakultät für Mathematik (MATH)
Institut Institut für Stochastik (STOCH)
Prüfungsdaten 18.05.2016
Referent/Betreuer Prof. T. Gneiting
Schlagworte Probabilistic forecasting, forecast evaluation, extreme events
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft KITopen Landing Page