Colloquium
The Department of Statistics
presents
 
Jan Mielniczuk
Polish Academy of Sciences
Warsaw, Poland

 

Some Results on Functional Estimation
for Long-Range Dependent Sequences
 
 
Abstract
 

Inferential methods for strongly dependent data receive much attention currently.  As the introduction, we argue that this happens not entirely without reason.  The main aim of the lecture is to analyze the behavior of certain long-range dependent sequences and curve estimators based on them.  We consider a strictly stationary long-range dependent process (Zi) with standard exponential marginals and the specific  bivariate densities as well as its subordinated process  (G(Zi )) for any square integrable function G.  We explain the reasons why the asymptotic behavior of partial-sum process of a long-range dependent sequence  (G(Zi)) is the same as that of the first non vanishing term of its Laguerre expansion (Lm(Zi)).  Furthermore, convergence in distribution of partial-sum process to a certain non-Gaussian process  is shown. This leads to it non-central  limit theorems for an empirical process and kernel density estimators.
 
 
 

Monday, February 9, 1998
4:10 P.M.,  1070 CEB (Duncan Hall)
4:00 P.M.: Coffee, 1044 CEB
 

 
 

back to Department of Statistics Home Page
This page last maintained on 02/02/98
Problems or suggestions to webmaster@stat.rice.edu
Contact the Department of Statistics via stat@stat.rice.edu