Arbeitspapier

Localizing strictly proper scoring rules

When comparing predictive distributions, forecasters are typically not equally interested in all regions of the outcome space. To address the demand for focused forecast evaluation, we propose a procedure to transform strictly proper scoring rules into their localized counterparts while preserving strict propriety. This is accomplished by applying the original scoring rule to a censored distribution, acknowledging that censoring emerges as a natural localization device due to its ability to retain precisely all relevant information of the original distribution. Our procedure nests the censored likelihood score as a special case. Among a multitude of others, it also implies a class of censored kernel scores that offers a multivariate alternative to the threshold weighted Continuously Ranked Probability Score (twCRPS), extending its local propriety to more general weight functions than single tail indicators. Within this localized framework, we obtain a generalization of the Neyman Pearson lemma, establishing the censored likelihood ratio test as uniformly most powerful. For other tests of localized equal predictive performance, results of Monte Carlo simulations and empirical applications to risk management, inflation and climate data consistently emphasize the superior power properties of censoring.

Sprache
Englisch

Erschienen in
Series: Tinbergen Institute Discussion Paper ; No. TI 2023-084/III

Klassifikation
Wirtschaft
Thema
Density forecast evaluation
Tests for equal predictive ability
Censoring
Likelihood ratio
CRPS

Ereignis
Geistige Schöpfung
(wer)
de Punder, Ramon
Diks, Cees G. H.
Laeven, Roger J. A.
van Dijk, Dick
Ereignis
Veröffentlichung
(wer)
Tinbergen Institute
(wo)
Amsterdam and Rotterdam
(wann)
2023

Handle
Letzte Aktualisierung
10.03.2025, 11:43 MEZ

Datenpartner

Dieses Objekt wird bereitgestellt von:
ZBW - Deutsche Zentralbibliothek für Wirtschaftswissenschaften - Leibniz-Informationszentrum Wirtschaft. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.

Objekttyp

  • Arbeitspapier

Beteiligte

  • de Punder, Ramon
  • Diks, Cees G. H.
  • Laeven, Roger J. A.
  • van Dijk, Dick
  • Tinbergen Institute

Entstanden

  • 2023

Ähnliche Objekte (12)