|
Activity Number:
|
389
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #305492 |
|
Title:
|
Spatial Multivariate EOFs: Discrete to Continuous Approximations
|
|
Author(s):
|
Yonggang Yao*+ and Noel Cressie
|
|
Companies:
|
The Ohio State University and The Ohio State University
|
|
Address:
|
1958 Neil Ave., Columbus, OH, 43210-1247,
|
|
Keywords:
|
discrete-to-continuous approximation ; empirical orthogonal function ; multivariate random field ; smoothing interpolation ; smoothing spline
|
|
Abstract:
|
Multivariate empirical orthogonal function analysis of multivariate spatial data has been practiced in many disciplines. Due to having a finite number of observations, people often have to extend the mEOF to the whole multivariate random field by using some discrete-to-continuous approximation (DCA) algorithm (e.g., smoothing interpolation). However, the advantages and disadvantages of DCAs have not been explored adequately. This paper considers model-based criteria for choosing a DCA. As an example, the smoothing-spline DCA is considered and applied to Iowa (1970--1990) temperature data.
|
