Naslov
On the tail index inference based on the scaling function method

Autori
Grahovac, Danijel ; Jia, Mofei ; Leonenko, Nikolai N. ; Taufer, Emanuele

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the 18th European Young Statisticians Meeting / Šuvak, Nenad - Osijek : J. J. Strossmayer University of Osijek, Department of Mathematics, 2014, 39-45

ISBN
978-953-6931-70-5

Skup
18th European Young Statisticians Meeting

Mjesto i datum
Osijek, Hrvatska, 26.08.2013. - 30.08.2013

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
heavy-tailed distributions; tail index; empirical structure function; scaling functions; Hill estimator

Sažetak
In Grahovac et al. (2014) arXiv:1310.0333, a new method has been presented for making inference about the tail of samples coming from unknown heavy-tailed distribution. Method is based on asymptotic properties of the empirical structure function, a variant of statistic that resembles usual sample moments. Using this approach one can successfully inspect the nature of the tail of the underlying distribution, as well as provide estimated values on the unknown tail index. Here we briefly describe the method and test its performance on some simulated and real world data by comparing it with the well known Hill estimator.

Izvorni jezik
Engleski

Znanstvena područja
Matematika

POVEZANOST RADA