On error bounds for high-dimensional asymptotic distribution of L 2 -type test statistic for equality of means

On error bounds for high-dimensional asymptotic distribution of L 2 -type test statistic for equality of means

Two new asymptotic approximations for the distribution of Chen and Qin's statistic are derived and their explicit error bounds are established. The proposed approximations are shown to be far more accurate than the conventional normal limits in large-p-small-n settings which is successfully app...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Statistics & probability letters 2020-02, Vol.157, p.108637, Article 108637
Hauptverfasser: Hyodo, Masashi, Nishiyama, Takahiro, Pavlenko, Tatjana
Format: Artikel
Sprache:eng
Schlagworte:
Chi-squared approximation
Convergence rate
Higher order Edgeworth expansion
Normal approximation
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Two new asymptotic approximations for the distribution of Chen and Qin's statistic are derived and their explicit error bounds are established. The proposed approximations are shown to be far more accurate than the conventional normal limits in large-p-small-n settings which is successfully approved by the numerical experiments.
ISSN:0167-7152
1879-2103
DOI:10.1016/j.spl.2019.108637