Bayesian Analysis for Population Ecology

King, Ruth; Morgan, Byron J. T.; Giménez, Olivier; Brooks, Stephen

doi:10.1201/9781439811887

Cited by 136 publications

(121 citation statements)

References 0 publications

Supporting

Mentioning

120

Contrasting

Unclassified

Order By: Relevance

“…For further discussion on this likelihood see for example Catchpole et al (1998), King and Brooks (2002), King et al (2009) andMcCrea andMorgan (2014).…”

Section: Notationmentioning

confidence: 99%

“…where α and β are the regression parameters (see for example North and Morgan 1979;King et al 2009). The set of model parameters is θ = { p, λ, α, β}, since the survival probabilities are a deterministic function of α and β.…”

Section: Modelmentioning

confidence: 99%

“…To incorporate additional individual heterogeneity in the survival probabilities, φ, we specify the survival probabilities as a function of individual time-varying continuous covariates (Catchpole et al 2000(Catchpole et al , 2004Bonner and Schwarz 2006;King et al 2006King et al , 2008King et al , 2009). We consider a single covariate, for simplicity, but the method can be immediately extended to multiple covariates (see Sect.…”

Section: Modelmentioning

confidence: 99%

“…Note that we use the term "recaptures" to include both physical recaptures and resightings. In the presence of unknown continuous covariate values, the corresponding likelihood is analytically intractable, and expressed as an integral over all the missing covariate values (see for example, Bonner and Schwarz 2006;King et al 2009;Langrock and King 2013 for further discussion). Several approaches have been proposed for dealing with this issue.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Analysing Mark–Recapture–Recovery Data in the Presence of Missing Covariate Data Via Multiple Imputation

Worthington

King

Buckland

2014

JABES

Self Cite

View full text Add to dashboard Cite

We consider mark-recapture-recovery data with additional individual time-varying continuous covariate data. For such data it is common to specify the model parameters, and in particular the survival probabilities, as a function of these covariates to incorporate individual heterogeneity. However, an issue arises in relation to missing covariate values, for (at least) the times when an individual is not observed, leading to an analytically intractable likelihood. We propose a two-step multiple imputation approach to obtain estimates of the demographic parameters. Firstly, a model is fitted to only the observed covariate values. Conditional on the fitted covariate model, multiple "complete" datasets are generated (i.e. all missing covariate values are imputed). Secondly, for each complete dataset, a closed form complete data likelihood can be maximised to obtain estimates of the model parameters which are subsequently combined to obtain an overall estimate of the parameters. Associated standard errors and 95 % confidence intervals are obtained using a non-parametric bootstrap. A simulation study is undertaken to assess the performance of the proposed twostep approach. We apply the method to data collected on a well-studied population of Soay sheep and compare the results with a Bayesian data augmentation approach.Supplementary materials accompanying this paper appear on-line.

show abstract

“…For further discussion on this likelihood see for example Catchpole et al (1998), King and Brooks (2002), King et al (2009) andMcCrea andMorgan (2014).…”

Section: Notationmentioning

confidence: 99%

Section: Modelmentioning

confidence: 99%

Section: Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Analysing Mark–Recapture–Recovery Data in the Presence of Missing Covariate Data Via Multiple Imputation

Worthington

King

Buckland

2014

JABES

Self Cite

View full text Add to dashboard Cite

show abstract

“…For clarity, other derived parameters (meaning they are functions of the estimated parameters) that are referred to in the methods are listed in Table 4. A Metropolis-Hastings Markov chain Monte Carlo (MCMC) algorithm (King et al, 2010) was used to generate samples of parameter vectors (and output quantities of interest) from the posterior distribution; these samples are then used to approximate the posterior distribution. This algorithm generates a random walk through the parameter space by selecting a test set of parameters from the neighborhood of the current parameter set, estimating the likelihood for the test parameter set, then applying a stochastic acceptance test to determine if the chain updated to the new parameters or remained at the current parameters.…”

Section: G Tmentioning

confidence: 99%