Huber estimation for the network autoregressive model

We propose the Huber estimator for the network autoregressive model, which exhibits both robustness and efficiency under heavy-tailed errors and loses little efficiency compared to OLS under light-tailed errors.

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Statistics & probability letters 2023-12, Vol.203, p.109917, Article 109917
Hauptverfasser: Xiao, Xuan, Xu, Xingbai, Zhong, Wei
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We propose the Huber estimator for the network autoregressive model, which exhibits both robustness and efficiency under heavy-tailed errors and loses little efficiency compared to OLS under light-tailed errors.
ISSN:0167-7152
1879-2103
DOI:10.1016/j.spl.2023.109917