|
Activity Number:
|
661
- Statistical Models for Animal Behavior and Population Dynamics
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 1, 2019 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
|
Abstract #304325
|
Presentation
|
|
Title:
|
An Irregular Sampling Design for Animal Movement
|
|
Author(s):
|
Elizabeth Eisenhauer* and Ephraim Hanks
|
|
Companies:
|
The Pennsylvania State University and Pennsylvania State University
|
|
Keywords:
|
Animal Movement;
Bayesian Statistics;
Stochastic Differential Equations;
Ecology
|
|
Abstract:
|
With rapid technological advances in telemetric devices, animal movement studies have become ubiquitous in animal ecology for estimation of space use and analysis of animal movement behavior. Animal telemetry data is primarily recorded at regularly spaced intervals in time, but we propose that an irregular sampling design could lead to greater efficiency and information gain in animal movement modeling. We present a novel sampling design called lattice and random intermediate point (LARI), which entails sampling at regular time intervals coupled with one randomly selected time point in between each adjacent pair of regular samples. We compare LARI to regularly spaced sampling schemes in an example with common black carpenter ant location data and a simulation with a quadratic potential function. We modify a stochastic differential equation model to allow for irregular intervals and fit the model on subsets of the data. Comparison of model parameter estimates and interpolation accuracy shows that LARI performs better than regularly-spaced sampling designs with regard to parameter estimation.
|
Authors who are presenting talks have a * after their name.
