Atlas home || Conferences | Abstracts | about Atlas
Host: University of Copenhagen
Homepage: http://www.stat.ku.dk/maphysto/empproc.html
Organizers: Thomas Mikosch, Søren Feodor Nielsen, Michael Sørensen
Description:
Empirical process techniques for independent and identically distributed observations have been used for many
years in statistics and probability theory. Over the last few years the need to model dependence in real-life data
has led to new developments for the empirical distribution function and the empirical process for dependent,
mostly stationary sequences. By now, the theory for dependent data is well understood, but the results are
spread widely in the literature. In addition to this fact, the classical empirical process techniques have been
adapted to various other fields, including extreme value theory and time series analysis. The tail empirical
process has become a useful tool for the statistics of extremal events of dependent data. It constitutes a link
between point processes and empirical processes. Similarly, the periodogram and its integrated versions are
analogues of the empirical process, and therefore ideas from that theory have been adapted to the spectral
analysis of time series.
The aim of the instructional workshop is to give an introduction to the existing theory of empirical process techniques for dependent data and to provide an overview of the most recent developments in various fields related to empirical processes. These include the spectral analysis of time series, the bootstrap for stationary sequences, the empirical process for mixing dependent observations, including the case of strong dependence. The introduction to the theory will be given in lectures by H. Dehling and W. Philipp. They will give the basics of the theory, starting form the classical contributions of P. Billingsley and covering the development up to the present. In addition to these lectures, some of the top specialists in the field will lecture on the most recent results in their areas.
Speakers: Herold Dehling (Ruhr-Universität Bochum), Walter Philipp (University of Illinois at Urbana-Champaign)
Date received: September 20, 2000
© 2008 Atlas Conferences Inc.