|
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 - #306984 |
|
Title:
|
Spatial Designs and Strength of Spatial Signal: Effects on Covariance Estimation
|
|
Author(s):
|
Kathryn Irvine*+ and Alix Gitelman and Jennifer A. Hoeting
|
|
Companies:
|
Oregon State University and Oregon State University and Colorado State University
|
|
Address:
|
1905 NW Dogwood Drive, Corvallis, OR, 97330,
|
|
Keywords:
|
exponential covariance ; nugget-to-sill ratio ; infill asymptotics
|
|
Abstract:
|
Interpretations of spatial covariance parameters in ecological settings can be used to describe the spatial heterogeneity in a landscape that cannot be explained by covariates. In this paper we investigate the influence of the strength of spatial signal on maximum likelihood (ML) and restricted maximum likelihood estimates (REML) of covariance parameters in an exponential with-nugget model, and we also examine these influences under different sampling designs---specifically, lattice designs and more realistic random and cluster designs---at differing intensities of sampling (n=144 and 361). We find that neither ML nor REML estimates perform well when the range is large, and that the best estimation of the covariance parameters comes under the random sampling design. ML underestimate the autocorrelation function and REML produces highly variable estimates of the autocorrelation function.
|
