Statistics of extremes under random censoring

Abstract

We investigate the estimation of the extreme value index when the data are subject to random censorship. We prove, in a unified way, detailed asymptotic normality results for various estimators of the extreme value index and use these estimators as the main building block for estimators of extreme quantiles. We illustrate the quality of these methods by a small simulation study and apply the estimators to medical data.

Published in at http://dx.doi.org/10.3150/07-BEJ104 the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm)

References in corpus (1)

• On maximum likelihood estimation of the extreme value index

Cited by in corpus (11)

• Modelling Censored Losses Using Splicing: a Global Fit Strategy With Mixed Erlang and Extreme Value Distributions
• Understanding heavy tails in a bounded world or, is a truncated heavy tail heavy or not?
• Tail empirical process and weighted extreme value index estimator for randomly right-censored data
• Estimation of the extreme value index in a censorship framework: asymptotic and finite sample behaviour
• Moment estimators of the extreme value index for randomly censored data in the Weibull domain of attraction
• Statistical estimate of the proportional hazard premium of loss under random censoring
• Logarithm of ratios of two order statistics and regularly varying tails
• Tails and probabilities for $p$-outside values
• A sequential design for extreme quantiles estimation under binary sampling
• The PORTSEA (Portuguese School of Extremes and Applications) and a few personal scientific achievements
• Estimating the mean of a heavy-tailed distribution under random censoring