Caley & Barry's (2014) "Non-constant" model

Non-constant population model from Caley & Barry 2014. Estimates a posterior probability that the species is extant at the test time, and a point estimate and one-sided \(1 - \alpha\) credible interval on the time of extinction.

Usage

CB14B2(
  records,
  alpha = 0.05,
  init.time = min(records),
  test.time = as.numeric(format(Sys.Date(), "%Y")),
  burn.in = 10000,
  n.iter = 110000
)

Arguments

records: sighting records in cbin format (see convert_dodo for details).
alpha: desired threshold level (defaults to \(\alpha = 0.05\)) of the \(1 - \alpha\) credible interval.
init.time: start of the observation period.
test.time: time point to retrospectively calculate extinction probability at. Defaults to the end of the observation period.
burn.in: number of initial iterations to discard as burn-in (defaults to 10,000).
n.iter: number of iterations to run (defaults to 110,000).

Value

a list object with the original parameters and the p(extant), point estimate, and credible interval included as elements. The credible interval is a two-element numeric vector called cred.int.

Note

All sighting records are assumed to be certain and sampling effort is assumed to be constant.

References

Key Reference

Caley, P., & Barry, S. C. (2014). Quantifying extinction probabilities from sighting records: inference and uncertainties. PLoS One, 9(4), e95857. doi:10.1371/journal.pone.0095857

Examples

# Run the fox analysis from Caley & Barry 2014
CB14B2(fox, init.time = 2001, test.time = 2012, n.iter = 11e4, burn.in = 1e4)
#> $records
#>  [1] 1 0 1 0 1 1 0 0 0 0 0 0
#> 
#> $alpha
#> [1] 0.05
#> 
#> $init.time
#> [1] 2001
#> 
#> $test.time
#> [1] 2012
#> 
#> $burn.in
#> [1] 10000
#> 
#> $n.iter
#> [1] 110000
#> 
#> $p.extant
#> [1] 0.29412
#> 
#> $estimate
#> [1] 2009
#> 
#> $cred.int
#> [1] 2007  Inf
#> 
if (FALSE) { # \dontrun{
# Run an example analysis using the Slender-billed Curlew data
CB14B2(curlew$cbin, init.time = 1817, n.iter = 11e3, burn.in = 1e3)
} # }