Extreme value statistics for censored data with heavy tails under competing risks

Julien Worms (1) , Rym Worms (2)

1 LMV - Laboratoire de Mathématiques de Versailles
2 LAMA - Laboratoire d'Analyse et de Mathématiques Appliquées

This paper addresses the problem of estimating the extreme value index in presence of random censoring and competing risks for heavy tailed distributions. Asymptotic normality of the proposed estimator (which has the form of an Aalen-Johansen integral) is established. A small simulation study exhibits its performances for finite samples.

Mots clés: Extreme value index; Tail inference; Random censoring; Competing Risks; Aalen-Johansen estimator

Domaines: Mathématiques [math]; Statistiques [stat]

Julien Worms, Rym Worms. Extreme value statistics for censored data with heavy tails under competing risks. 2016. ⟨hal-01418370v1⟩