******************************************************************************

                                    Seminar


       Department of Systems Engineering and Engineering Management,

                    The Chinese University of Hong Kong

Title:
Inference for Nearly Unstable Processes under Strong Dependence

Speaker:
Prof. Ngai Hung CHAN
Department of Statistics
Chinese University of Hong Kong

Date : February 10, 2006 (Friday)

Time : 4:30 p.m. - 5:30 p.m.

Venue : Room 513, William M.W. Mong Engineering Building

(Engineering Building Complex Phase 2), CUHK

Abstract:
This talk considers the effect of least squares procedures for nearly unstable
linear time series when innovations are strongly dependent. We compare our
results to those dealing with uncorrelated innovations. Our analysis includes
fractional integrated noise and fractional Gaussian noise. Under appropriate
scaling, we show that ordinary least squares procedures converge to
functionals of fractional Ornstein-Uhlenbeck processes. The functionals bear
only formal analogy to those in the classical framework with uncorrelated
innovations. It will be shown that limit theorems for the functionals involve
non-standard scaling and non-standard limiting distributions. Results of this
paper shed light on the asymptotic behaviors of nearly unstable long-memory
processes.

Bio:
Professor N.H. Chan is the Chairman and Professor of Statistics at CUHK. He
was the Director of the Risk Management Science program of CUHK before
becoming the Chairman of the Statistics Department. He received his
undergraduate training at CUHK and his Ph.D. from the University of Maryland.
Before returning to CUHK, he was Professor of Statistics at Carnegie Mellon
University and Indiana University. His research interests include time series,
econometrics, risk management and stochastic processes.

Note : Cookies and drinks will be available at 4:15 pm.

 _______________________________________________________________________________