simmr_clone: Clone a data-frame then run a simmr model
In cbrown5/remixsiar: Post-processing of stable isotope mixing models

Description Usage Arguments Details Value Author(s)

Data cloning can be used to obtain maximum-likelihood estimates of parameters from Bayesian MCMC models. See details.

1	simmr_clone(dat, K = 20, mcmc.control, prior.control)

`dat`	A `list` containing the inputs for `simmr::simmr_load` including the elements named `mixes`, `s_names`, `source_means`, `source_sds`.
`K`	A `numeric` integer giving the number of data-clones.
`mcmc.control`	A `list` specifying the parameters for the mcmc algorithm, including `iter`, `burn`, `thin`, `n.chain`. See the `simmr` documentation for more details.
`prior.control`	A `list` specifying the parameters for the models' priors, including `means` and `sds`. To generate default priors see `default_prior`

Data cloning is a legitimate method for obtaining maximum-likelihood estimates of parameters from Bayesian models. This may be useful for estimating prior influence (see remix_shrink) or if frequentist statistcs are desired instead of Bayesian statistics.

It is very important to choose a large enough value of K. It is recommended to run cloning with various K values and check parameter estimates have converged.

This algorithm may take considerable time to run if your data-set is large.

For more details, see: Lele SR, Dennis B, Lutscher F. Data cloning: easy maximum likelihood estimation for complex ecological models using Bayesian Markov chain Monte Carlo methods. Ecology letters. 2007 Jul 1;10(7):551-63.

A simmr::simmr_output object. See the documentation of that package for more details.

Christopher J. Brown christo.j.brown@gmail.com

cbrown5/remixsiar documentation built on April 26, 2020, 12:40 a.m.