Activity Number:
|
183
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 13, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Computing*
|
Abstract - #301252 |
Title:
|
On the Use of Independent Component Analysis to Separate Meaningful Sources in Global Temperature Series
|
Author(s):
|
Imola Fodor*+ and Chandrika Kamath
|
Affiliation(s):
|
Lawrence Livermore National Laboratory and Lawrence Livermore National Laboratory
|
Address:
|
P.O. Box 808, L-560, Livermore, California, 94551,
|
Keywords:
|
Statistical computing ; Independent component analysis ; Time series analysis ; Climate data
|
Abstract:
|
Global temperature series have contributions from different sources, such as volcanic eruptions and El Ni\~no/Southern Oscillation variations. We investigate independent component analysis (ICA) as a technique to separate unrelated sources present in such series. We first use artificial data, with known independent components, to study the conditions under which ICA separates the individual sources. We then illustrate the method with climate data from the National Centers for Environmental Prediction.
|