|
Activity Number:
|
5
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, August 8, 2004 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #300160 |
|
Title:
|
State-space Models for Biological Monitoring Data
|
|
Author(s):
|
Devin S. Johnson*+ and Jennifer A. Hoeting
|
|
Companies:
|
University of Alaska, Fairbanks and Colorado State University
|
|
Address:
|
Dept. of Mathematical Sciences, Fairbanks, AK, 99775-6660,
|
|
Keywords:
|
compositional data ; graphical models ; logistic normal ; random effects ; species assemblage ; conditional independence
|
|
Abstract:
|
An emerging area of research in ecology is the analysis of functional species assemblages. In essence, the analysis of functional assemblages is concerned with determining and predicting the composition of individuals categorized using different life history traits instead of strict taxa names. We propose a state-space model for the analysis of multiple trait compositions along with site-specific covariate information. A site-specific random effects term allows for modeling extra variability including spatial variability in trait compositions. This approach has several advantages over the traditional logistic normal model used in the analysis of similar compositional data. The model can also be considered in terms of a chain graph model. If there are no structural zeros in the space of possible trait combinations (combinations of traits that are impossible), we show that the model parameters correspond to conditional independence relationships. Using a Gibbs sampling approach, we illustrate application of the model on a data set of fish species richness in the mid-Atlantic region of the U.S.
|
