R: Draw a random model-fit from the sampling distribution

emulate {ctmm}

R Documentation

Draw a random model-fit from the sampling distribution

Description

This function generates random model-fit statistics from the sampling distribution of a given ctmm movement model and sampling schedule. If fast=FALSE, the results are exact, though slow to evaluate. Else if fast=TRUE, the central-limit theorem is invoked.

Usage

emulate(object,...)

## S3 method for class 'ctmm'
emulate(object,data=NULL,fast=FALSE,...)

## S3 method for class 'telemetry'
emulate(object,CTMM,fast=FALSE,...)

Arguments

`object`	`telemetry` data or `ctmm` model object.
`CTMM`	A `ctmm` movement-model object.
`data`	Optional `telemetry` object for exact results.
`fast`	Whether or not to invoke the central-limit theorem.
`...`	Arguments passed to `ctmm.fit`.

Details

Given fast=FALSE, which requires the data argument specified, new data are simulated from the CTMM movement model with the same sampling schedule and error structure as data. A new model, of the same structure as CTMM, is then fit to the simulated data and returned.

Given fast=TRUE, a model-fit object is sampled from the central-limit distribution, using the covariance estimates within CTMM. Strictly positive parametes, such as area, are log-transformed prior to the normal approximation. Note that this faster method does not adjust for bias.

Value

A ctmm movement model with the same structure as CTMM.

Author(s)

C. H. Fleming.