Summarize simulated exposures relative to reference subject
forestSummarize(data, funs.exposure, cover.ci = 0.95, by, as.fun)A data.frame
Simulated data to process. This data.frame must contain must contain multiple columns, as defined by NMsim::expandCovs().
A named list of functions to apply for derivation of exposure metrics.
The coverage of the confidence intervals. Default is 0.95.
a character vector of column names to perform all calculations by. This could be sampling subsets or analyte.
The default is to return data as a `data.frame`. Pass a function (say `tibble::as_tibble`) in as.fun to convert to something else. If data.tables are wanted, use `as.fun="data.table"`. The default can be configured using `NMdataConf()`.
This function is part of the workflow provided by NMsim to generate forest plots - a graphical representation of the estimated covariate effects and the uncertainty of those effect estimates. `expandCovs()` helps construct a set of simulations to perform, simulation methods like `NMsim_VarCov` and `NMsim_NWPRI` can perform siulations with parameter uncertainty, and `forestSummarize()` can then summarize those simulation results into the numbers to plot in a forest plot. See the NMsim vignette on forest plot generation available on the NMsim website for a step-by-step demonstration.
The following columns are generated by `expandCovs()` and are expected to be present. Differences within any of them will lead to separate summarizing (say for as covariate value to be plotted):
`model`: A model identifier - generated by `NMsim()`.
`type`: The simulation type. "ref" for reference subject, "value" for any other. This is generated by `expandCovs()`.
`covvar`: The covariate (of interest) that is different from the reference value in the specific simulation. Example: "WT"
`covlabel`: Label of the covariate of interest. Example: "Bodyweight (kg)"
`covref`: Reference value of the covariate of interest. Example: 80
`covval`: Value of the covariate of interest (not reference). Example 110.