simdata: Simulate consumer isotope ratios to evaluate posterior model...
In cbrown5/remixsiar: Post-processing of stable isotope mixing models

Description Usage Arguments Details Value Author(s)

Simulations of consumer isotope ratios are useful to evaluate model fits for bias and precision (variance). This function simulates data from simmr_output objects to estimate model bias and precision. Multiple groups are allowed.

1	simdata(simmr_in, simmr_out, fast = TRUE)

`simmr_in`	A `simmr_input` object, the output of simmr_mcmc
`simmr_out`	A `simmr_output` object
`fast`	A `logical` indicating whether to sub-sample the full mcmc chain, to speed simulation of new data.

High bias is indicative of a mis-specified model, for instance one that is missing an important source or an inaccurate fractionation. See remix_PI and remix_2DPI for plotting model predictions for 1 and 2 tracers.

A remix_simdata object containing for each tracer:

`bias`	A `data.frame` with estimates of model bias - mean deviation of the consumer data from the mean estimate
`rmse`	Variance of consumer isotope ratio predictions estimated as root-mean-square error.
`xmean`	Mean predicted consumer isotope ratio
`xvar`	Variance in predicted consumer isotope ratios