Estimation of the extreme value index and extreme quantiles under random censoring (Q1003306): Difference between revisions

The random right censoring model is considered where the observations and the censor belong to domains of max-attraction of extreme value distributions. The problem is to estimate the extreme value index of the observations. Maximum likelihood and adapted moment estimates are considered. Asymptotic normality of the moment estimate and corresponding extreme quantile estimate are demonstrated. Results of simulations and applications to medical data are presented.