simmr_fitsims: Fits a simmr 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

simmr_fitsims(n, tracer_means, tracer_sds, snames, smeans, ssds, sdmult = 1,
  seed = 42, mcmc.control = list(iter = 50000, burn = 2000, 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.
`smeans`	A `numeric` vector giving means of the sources
`ssds`	A `numeric` vector giving SDs of the mixes
`sdmult`	A `numeric` that is multiplied by source SDs
`mcmc.control`	A list giving parameters for the MCMC chains.
`sdmult`	A `numeric` giving the random seed
`priorcontrol`	A list with the means and standard deviations of the priors.

The mixes and sampled randomly from normal distributions. Make sure the means and SDs of mixes and sources are given in the same order (e.g. Carbon then Nitrogen for all of them). Fixing the seed is useful if you want to compare runs across different SDs, because the random samples will have the same pattern deviations. See simmr::simmr_load simmr::simmr_mcmc for more details of how the data is fit.

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

Christopher J. Brown

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