|
Activity Number:
|
326
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #308172 |
|
Title:
|
Modeling Unobserved Sources of Heterogeneity in Animal Abundance Using a Dirichlet Process Prior
|
|
Author(s):
|
Robert Dorazio*+ and Bhramar Mukherjee and Li Zhang and Malay Ghosh
|
|
Companies:
|
U.S. Geological Survey/University of Florida and University of Michigan and The Cleveland Clinic and University of Florida
|
|
Address:
|
Dept of Statistics, Gainesville, FL, 32611-0339,
|
|
Keywords:
|
abundance estimation ; detection heterogeneity ; Dirichlet process ; Bayesian nonparametric
|
|
Abstract:
|
In surveys of natural animal populations, a sampling protocol is often spatially replicated to collect a representative sample of the population. In these surveys differences in abundance of animals among sample locations may induce spatial heterogeneity in the counts associated with a particular sampling protocol. For some species, the sources of heterogeneity in abundance may be unknown or immeasurable, leading one to specify the variation in abundance among sample locations stochastically. However, choosing a parametric model for the distribution of unmeasured heterogeneity is potentially subject to error and can have profound effects on predictions of abundance at unsampled locations. We develop a robust alternative approach wherein a Dirichlet process prior is assumed for the distribution of latent abundances.
|
