A0189
Title: Asymptotics of peaks-over-threshold estimators in long memory linear time series
Authors: Marco Oesting - University of Stuttgart (Germany) [presenting]
Gilles Stupfler - University of Angers (France)
Ioan Scheffel - University of Stuttgart (Germany)
Abstract: The purpose is to consider peaks-over-threshold (POT) estimators for extremes of long memory linear time series. As these time series are not beta-mixing, classical asymptotic results on POT estimators are not applicable. A reduction principle for subordinated long memory linear time series is adapted to this setting. Thus, a central limit theorem is proven for POT estimators, including the Hill estimator. Convergence to stable limit distributions is obtained with different rates for light and heavy tails.
