Abstract: In the context of band recovery models with heterogeneous survival rates, this paper estimates annual average survival rates by estimating the size of surviving population of banded birds. For the case of homogeneous survival rates, Brownie, Anderson, Burnham and Robson (1985) derived the relationship between the total number of surviving banded birds and annual average survival rates. However, their estimator of the total number of surviving banded birds is biased in the case of heterogeneous survival rates. We generalize their result to the case that annual average survival rates vary across individuals and years. In addition, the coefficient of variation for individual survival rates is used to reduce the estimation bias. The analytically intractable variances of proposed estimators are obtained by the bootstrap method. The proposed method is applied to data, and a simulation study is conducted to compare the performances of the estimators. Numerical results indicate that the proposed method works satisfactorily in general, and when there is a high degree of heterogeneity in particular.
Key words and phrases: Band recovery model, bootstrap sampling, coefficient of variation, heterogeneous survival rate, maximum likelihood estimation.