Study of K-Sample Capture-Recapture Experiments with Known Losses on Capture Under Partial Stratification

• Activity Number: 31
• Type: Contributed
• Date/Time: Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Lasantha Premarathna* and Carl Schwarz
• Companies: Simon Fraser University and Simon Fraser University
• Keywords: Abundance ; Bayesian analysis ; Capture heterogeneity ; Loss on capture ; MLE ; Survey-design and analysis

Abstract

Analysis of capture-recapture data under closed population models assumes the demographical and geographical closure. Capture-recapture studies with known losses on capture (some captured animals at each sample time may not be available for the recapture at subsequent sample times) are still considered under the closure assumption. Capture heterogeneity cause bias in estimate of abundance in these types of studies. Heterogeneity is related to observable fixed characteristic of the animal such as sex. If the each sampled animal can be stratified, then it is straightforward to obtain stratum-specific estimates. In many experiments it is difficult to stratify all the captures animals. In these cases a sub-sample of the captured animals at each sampling time is selected. In this study, we develop and apply new methods for these types of experiments using MLE methods. There are costs associate with capture an animal and processing the sub-sample at each sample time. We further determine the optimal allocation of effort for a given cost. Also We develop methods using Bayesian approach. An illustrative example is presented by applying these new methods to simulated data.