|
Activity Number:
|
13
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #310416 |
|
Title:
|
Nonparametric Modeling of the Second Order Structure of Processes with Time-Varying Memory
|
|
Author(s):
|
Georg Goerg*+ and Dana Draghicescu
|
|
Companies:
|
Vienna University of Technology and Hunter College
|
|
Address:
|
Argentinierstrabe 8/105, Vienna, 1040, , Austria
|
|
Keywords:
|
Hurst parameter ; nonstationary time series ; time-varying memory ; optimal bandwidth selection ; returns ; river flow
|
|
Abstract:
|
In this paper we consider nonstationary time series, characterized by time-varying memory. We model the temporal changes of the memory parameter (or Hurst parameter) in the time domain, by using a moving window approach. Our estimation procedure incorporates a data-driven scheme for optimal bandwidth selection. The proposed methodology is illustrated on hydrological and financial datasets.
|
