EUVIMEDEUVIMED entdecken
Crossref

Efficient functional estimation and the super-oracle phenomenon

Thomas B. Berrett, Richard J. Samworth

The Annals of Statistics · 01.04.2023 · 51 (2) · 668-690

DOI: 10.1214/23-aos2265 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Annals of Statistics, erschienen am 01.04.2023. Verfasst von Thomas B. Berrett und 1 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 33 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We consider the estimation of two-sample integral functionals, of the type that occur naturally, for example, when the object of interest is a divergence between unknown probability densities. Our first main result is that, in wide generality, a weighted nearest neighbour estimator is efficient, in the sense of achieving the local asymptotic minimax lower bound. Moreover, we also prove a corresponding central limit theorem, which facilitates the construction of asymptotically valid confidence intervals for the functional, having asymptotically minimal width. One interesting consequence of our results is the discovery that, for certain functionals, the worst-case performance of our estimator may improve on that of the natural ‘oracle’ estimator, which itself can be optimal in the related problem where the data consist of the values of the unknown densities at the observations.

Zitieren

APA

Berrett, T. B., & Samworth, R. J. (2023). Efficient functional estimation and the super-oracle phenomenon. The Annals of Statistics, 51(2), 668-690. https://doi.org/10.1214/23-aos2265

Vancouver

Berrett TB, Samworth RJ. Efficient functional estimation and the super-oracle phenomenon. The Annals of Statistics. 2023;51(2):668-690. doi:10.1214/23-aos2265

BibTeX
@article{Berrett2023,
  author = {Berrett, Thomas B. and Samworth, Richard J.},
  title = {Efficient functional estimation and the super-oracle phenomenon},
  journal = {The Annals of Statistics},
  year = {2023},
  volume = {51},
  number = {2},
  pages = {668--690},
  doi = {10.1214/23-aos2265}
}
Kostenlos starten →