Buchumschlag

Model diagnostics and forecast evaluation for quantiles

Model diagnostics and forecast evaluation are closely related tasks, with the former concerning in-sample goodness (or lack) of fit and the latter addressing predictive performance out-of-sample. We review the ubiquitous setting in which forecasts are cast in the form of quantiles or quantile-bounde...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gneiting, Tilmann (VerfasserIn) , Wolffram, Daniel (VerfasserIn) , Resin, Johannes (VerfasserIn) , Kraus, Kristof (VerfasserIn) , Bracher, Johannes (VerfasserIn) , Dimitriadis, Timo (VerfasserIn) , Hagenmeyer, Veit (VerfasserIn) , Jordan, Alexander I. (VerfasserIn) , Lerch, Sebastian (VerfasserIn) , Phipps, Kaleb (VerfasserIn) , Schienle, Melanie (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: March 2023
In: Annual review of statistics and its application
Year: 2023, Jahrgang: 10, Pages: 597-621
ISSN:2326-831X
DOI:10.1146/annurev-statistics-032921-020240
Online-Zugang:Verlag, kostenfrei, Volltext: https://doi.org/10.1146/annurev-statistics-032921-020240
Volltext
Verfasserangaben:Tilmann Gneiting, Daniel Wolffram, Johannes Resin, Kristof Kraus, Johannes Bracher, Timo Dimitriadis, Veit Hagenmeyer, Alexander I. Jordan, Sebastian Lerch, Kaleb Phipps, and Melanie Schienle
Beschreibung
Zusammenfassung:Model diagnostics and forecast evaluation are closely related tasks, with the former concerning in-sample goodness (or lack) of fit and the latter addressing predictive performance out-of-sample. We review the ubiquitous setting in which forecasts are cast in the form of quantiles or quantile-bounded prediction intervals. We distinguish unconditional calibration, which corresponds to classical coverage criteria, from the stronger notion of conditional calibration, as can be visualized in quantile reliability diagrams. Consistent scoring functions—including, but not limited to, the widely used asymmetricpiecewise linear score or pinball loss—provide for comparative assessment and ranking, and link to the coefficient of determination and skill scores. We illustrate the use of these tools on Engel's food expenditure data, the Global Energy Forecasting Competition 2014, and the US COVID-19 Forecast Hub.
Beschreibung:Zuerst veröffentlicht als "Review in Advance" am 01. November 2022
Gesehen am 26.05.2023
Beschreibung:Online Resource
ISSN:2326-831X
DOI:10.1146/annurev-statistics-032921-020240

Ähnliche Einträge