mixsiar_fitsims: Fits a MixSIAR model to a randomly generated data-set
In cbrown5/remixsiar: Post-processing of stable isotope mixing models

Description Usage Arguments Details Value Author(s)

Useful for obtaining fits from a sequence of randomly simulated datasets. For instance, you could use this function to perform a sort of broad-sense power analysis to ask how many samples are required to obtain a specific confidence about an animal's diet.

1
2
3

mixsiar_fitsims(n, tracer_means, tracer_sds, snames, modelfile, mixsiardat,
  sdmult = 1, seed = 42, mcmc.control = list(iter = 50000, burn = 10000,
  thin = 10, n.chain = 4), prior.control = NULL)

`n`	A `integer` giving the sample size to simulate.
`tracer_means`	A `numeric` vector giving means of the mixes.
`tracer_sds`	A `numeric` vector giving SDs of the mixes.
`snames`	A `character` giving source names.
`modelfile`	A `character` giving the location of the MixSIAR model filename.
`mixsiardat`	A `list` giving MixSIAR input data that will be modified for the re-run (should include objects mix, source, and disc, see the MixSIAR manual for further info).
`sdmult`	A `numeric` that is multiplied by source SDs
`mcmc.control`	A list giving parameters for the MCMC chains.
`priorcontrol`	A list with the means and standard deviations of the priors.

The mixes and sampled randomly from normal distributions. Fixing the seed is useful if you want to compare runs across different SDs, because the random samples will have the same pattern deviations.

A list where the first object gives the simulated data and the second object is a rjags object giving the model fit.

Christopher J. Brown

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