Catchpole, Edward A., Kgosi, P.M., Morgan, Byron J. T. (2001) On the near-singularity of models for animal recovery data. Biometrics, 57 (3). pp. 720-726. ISSN 0006-341X. (doi:10.1111/j.0006-341X.2001.00720.x) (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided) (KAR id:678)
The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided. | |
Official URL: http://dx.doi.org/10.1111/j.0006-341X.2001.00720.x |
Abstract
Certain probability models sometimes provide poor descriptions when fitted to data by maximum likelihood. We examine one such model for the survival of wild animals, which is fitted to two sets of data. When the model behaves poorly, its expected information matrix, evaluated at the maximum likelihood estimate of parameters, has a 'small' smallest eigenvalue. This is due to the fitted model being similar to a parameter-redundant submodel. In this case, model parameters that are precisely estimated have small coefficients in the eigenvector corresponding to the smallest eigenvalue. Approximate algebraic expressions are provided for the smallest eigenvalue. We discuss the general applicability of these results.
Item Type: | Article |
---|---|
DOI/Identification number: | 10.1111/j.0006-341X.2001.00720.x |
Uncontrolled keywords: | design; information matrix; near singularity; parameter redundancy; perturbation approximation; ring recovery data; smallest eigenvalue |
Subjects: | H Social Sciences > HA Statistics |
Divisions: | Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Mathematics, Statistics and Actuarial Science |
Depositing User: | Judith Broom |
Date Deposited: | 19 Dec 2007 18:25 UTC |
Last Modified: | 16 Nov 2021 09:39 UTC |
Resource URI: | https://kar.kent.ac.uk/id/eprint/678 (The current URI for this page, for reference purposes) |
- Export to:
- RefWorks
- EPrints3 XML
- BibTeX
- CSV
- Depositors only (login required):