|
Activity Number:
|
33
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #307189 |
|
Title:
|
On Nonparametric Smoothing Methods for Assessing Climate Change
|
|
Author(s):
|
Patricia Menendez Galvan*+ and Sucharita Ghosh
|
|
Companies:
|
Swiss Federal Research Institute WSL/ETHZ and Swiss Federal Research Institute WSL
|
|
Address:
|
Swiss Federal Research Institute WSL, Birmensdorf, CH-8903, Switzerland
|
|
Keywords:
|
curve estimation ; climate change ; time series ; PALAEO research
|
|
Abstract:
|
The Greenland Ice Core Project (Johnsen et al. 1997) provided data on past environmental conditions of the earth covering the last 250000 years. Such palaeo data are obtained from environmental proxies. For instance, oxygen isotopes trapped in ice are used to estimate past temperature. These long-term records reveal major environmental fluctuations and long-term shifts. Given the scenarios of arbitrary changes in the underlying process, it seems reasonable to use nonparametric smoothing methods to assess different aspects of climate change. In this paper, we expand on the ideas of Ghosh, Beran & Innes (1997) who allow the underlying (conditional) pdf to change over time. We consider one-dimensional transformations of Gaussian processes and examine their ramifications in the context of climate change. We obtain asymptotic results and illustrate our method via examples.
|
