Selection of Models of Lagged Identification Rates and Lagged Association Rates Using AIC and QAIC

The lagged identification rate is the probability of identifying an individual given its identification some time lag earlier. The lagged association rate is the probability that two individuals are associated given their association some time lag earlier. Models of lagged identification and association rates fit by maximizing the sums of non independent log-likelihoods have approximately unbiased parameter estimates. Simulations suggest that: Akaike-Information-Criterion often selects the true model of lagged identification rate data; quasi-Akaike-Information-Criterion performs better for lagged association rates; and confidence intervals for parameters are best obtained by bootstrap methods for lagged identification rates and quasi-likelihood or jackknife methods for lagged association rates.

[1]  H. Whitehead,et al.  Behavioural evidence for social units in long-finned pilot whales , 2003 .

[2]  Hal Whitehead,et al.  Investigating structure and temporal scale in social organizations using identified individuals , 1995 .

[3]  M. Woodroofe.  On Model Selection and the ARC Sine Laws , 1982 .

[4]  B. Würsig,et al.  Spinner dolphins in a remote Hawaiian atoll: social grouping and population structure , 2005 .

[5]  David R. Anderson,et al.  Model Selection and Multimodel Inference , 2003 .

[6]  Ray Hilborn,et al.  Determination of Fish Movement Patterns from Tag Recoveries using Maximum Likelihood Estimators , 1990 .

[7]  K. Burnham,et al.  Program MARK: survival estimation from populations of marked animals , 1999 .

[8]  Rick L. Andrews,et al.  Retention of latent segments in regression-based marketing models , 2003 .

[9]  J. Norris,et al.  Capture-Recapture Models with Heterogeneity : I . Cormack-Jolly-Seber Model , 2003 .

[10]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[11]  Hal Whitehead,et al.  Analysis of animal movement using opportunistic individual identifications: application to sperm whales , 2001 .

[12]  L. Bejder,et al.  A method for testing association patterns of social animals , 1998, Animal Behaviour.

[13]  S. Hooker,et al.  Population size and residency patterns of northern bottlenose whales (Hyperoodon ampullatus) using the Gully, Nova Scotia , 2000, J. Cetacean Res. Manage..

[14]  David R. Anderson,et al.  Model selection and multimodel inference : a practical information-theoretic approach , 2003 .

[15]  B. Efron,et al.  The Jackknife Estimate of Variance , 1981 .