On Tests to Distinguish Distribution Tails Invariant with Respect to the Scale Parameter
- 1 April 2022
- journal article
- research article
- Published by Pleiades Publishing Ltd in Doklady Mathematics
- Vol. 105 (2) , 97-101
Abstract
We propose two tests to distinguish between separable classes of distribution tails, the first of which is invariant with respect to the scale parameter and the second is invariant with respect to both location and scale parameters. The asymptotic properties of the proposed tests are established. The distributions are not assumed to belong to any maximum domain of attraction.Keywords
This publication has 19 references indexed in Scilit:
- Approximation of high quantiles from intermediate quantilesExtremes, 2016
- Approximation and estimation of very small probabilities of multivariate extreme eventsExtremes, 2016
- Extreme Value Theory and Statistics of Univariate Extremes: A ReviewInternational Statistical Review, 2014
- Weibull tail-distributions revisited: A new look at some tail estimatorsJournal of Statistical Planning and Inference, 2011
- Goodness-of-fit testing for Weibull-type behaviorJournal of Statistical Planning and Inference, 2010
- Review of testing issues in extremes: in honor of Professor Laurens de HaanExtremes, 2008
- Estimating Extreme Quantiles of Weibull Tail DistributionsCommunications in Statistics - Theory and Methods, 2005
- Second-order regular variation and rates of convergence in extreme-value theoryThe Annals of Probability, 1996
- The mean residual life function at great age: Applications to tail estimationJournal of Statistical Planning and Inference, 1995
- On the estimation of the Weibull tail coefficientJournal of Statistical Planning and Inference, 1993