|
Activity Number:
|
512
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #310183 |
|
Title:
|
Bias of Cormack-Jolly-Seber (CJS) Survival Estimators in Release Recapture Studies with Low Sampling Effort
|
|
Author(s):
|
James R. Faulkner*+ and Steven G. Smith
|
|
Companies:
|
National Marine Fisheries Service and NOAA Fisheries
|
|
Address:
|
2725 Montlake Blvd East, Seattle, WA, 98112-2097,
|
|
Keywords:
|
estimator bias ; release-recapture ; Cormack-Jolly-Seber ; sampling ; survival
|
|
Abstract:
|
CJS maximum likelihood (ML) estimators for survival probabilities and their associated variances are nearly asymptotically unbiased, but small effective sample sizes can result in substantial bias. Release-recapture experiments that follow a single marked release with low sampling effort are particularly susceptible to estimator bias. We use computer simulation to investigate the bias of CJS ML estimators and some bias-adjusted estimators of survival under conditions of low sampling effort. We also investigate the bias of the large sample variance of the survival estimators under various conditions.
|
