|
Activity Number:
|
201
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #306437 |
|
Title:
|
Spatio-Temporal Analysis Incorporating a Spatial Correlation Structure on a Long-Term Forestry Field Research Dataset
|
|
Author(s):
|
Bronson Bullock*+ and Edward Boone
|
|
Companies:
|
North Carolina State University and The University of North Carolina at Wilmington
|
|
Address:
|
3102 Jordan Hall, Department of Forestry & Environmental Research, Raleigh, NC, 27695-8008,
|
|
Keywords:
|
loblolly pine ; Pinus taeda ; repeated measures ; dynamic linear model
|
|
Abstract:
|
The use of spatio-temporal data sets is growing in environmental research fields. These data sets are often plagued with both spatial and temporal correlation. Often this data is collected at regular intervals and at predefined locations. A spatio-temporal model is presented using a Dynamic Linear Model with a parametric spatial correlation structure. In this paradigm the spatial correlation structure can be placed either on the mean or on the observation errors. Examples of data generated by each type of process are considered. This paradigm is applied to a loblolly pine tree (Pinus taeda) data set from a long-term forestry field research site.
|
