remix_PI: Estimate predictive intervals for consumer isotope ratios
In cbrown5/remixsiar: Post-processing of stable isotope mixing models

Description Usage Arguments Details Value Author(s)

Plotting the consumer's isotope values against those predicted by the model can be useful to detect model bias, or instance when consumers fall outside the source polygon. This function also returns residuals for the consumers as their deviations from the mean predicted value.

1 2	remix_PI(simdat, simmr_in, simmr_out, groups = NULL, plotresid = TRUE, probs = c(0.025, 0.975))

`simmr_in`	A `simmr_input` object.
`groups`	A `numeric` with the groups to plot if they are used, otherwise `NULL`.
`plotresid`	A `logical` which determines whether credibility residual plots are created.
`probs`	A `numeric` vector of two numbers giving the upper and lower predictive intervals.
`simdata`	A `remix_PI` object.
`simmr_in`	A `simmr_output` object.

New samples should fall within the predictive intervals with 95 Model bias is indicated by consumer data that fall outside the predictive intervals. Bias may occur, among other things due to missed sources or incorrect fractionation estimates.

A remix_PI object with