bootstrapCapwire: Parametric bootstrap ECM and TIRM models

In mwpennell/capwire: Estimates population size from non-invasive sampling

Parametric bootstrap ECM and TIRM models

Description

Uses maximum likelihood parameter estimates from fitEcm, or fitTirm to perform a parametric bootstrap to get confidence intervals for the estimate of the population size

Usage

bootstrapCapwire(fit, bootstraps = 1000, CI = c(0.025, 0.975))

Arguments

fit	a capfit object inherited from fitEcm or fitTirm
bootstraps	The number of bootstraps to be performed (default is 1000)
CI	A vector of quantiles to to generate a confidence interval for the population estimate. The default is c(0.025, 0.975), denoting a symmetrical 95 percent confidence interval.

Details

This function uses the ML estimates obtained from fitting the model to simulate data under the model.

bootstrapCapwire inherits an object from fitEcm or fitTirm such that the model and parameter estimates do not need to be specified.

The ML estimate for the population size will also be returned but this will not be changed by bootstrapCapwire

Note that if the model is a poor fit to the data, the confidence intervals may not be reliable.

The lower confidence interval is bounded by the number of unique individuals in the sample.

Value

A list containing the following elements:

ml.pop.size

The maximum likelihood estimate for the population size obtained by fitting the model

conf.int

The confidence interval for the estimate of the population size

Author(s)

Matthew W. Pennell

References

Miller C. R., P. Joyce and L.P. Waits. 2005. A new method for estimating the size of small populations from genetic mark-recapture data. Molecular Ecology 14:1991-2005.

Pennell, M. W., C. R. Stansbury, L. P. Waits and C. R. Miller. 2013. Capwire: a R package for estimating population census size from non-invasive genetic sampling. Molecular Ecology Resources 13:154-157.

See Also

fitTirm, fitEcm

Examples

## Simulate data under Equal Capture Model

data <- simEcm(n=40, s=150)

## Fit Equal Capture Model to Data

res <- fitEcm(data=data, max.pop=200)

## Perform bootstrap to get confidence intervals

ci <- bootstrapCapwire(fit=res, bootstraps=50, CI = c(0.025, 0.975))
ci

mwpennell/capwire documentation built on Jan. 12, 2023, 11:19 a.m.