Produces plots from a funmediation object produced by the funmediation function. These plots will be shown on the default output device (likely the screen); they can of course be written to a file instead, by preceding the call to plot with a call to png(), pdf(), or other R graphic file output functions.
# S3 method for funmediation
plot(
x,
use_panes = TRUE,
what_plot = c("pfr", "pfrgam", "coefs", "tvem"),
alpha_level = 0.05,
...
)This function does not return an object, but is called for its side effect of plotting to the active device.
The funmediation object to be plotted.
Whether to plot multiple coefficient functions in a single image.
One of "pfr","coefs", or "tvem." These options are as follows:
For a "pfr" plot, the functional coefficient for predicting the outcome, Y, from the mediator, M (conditional on X), is shown.
Similar to a "pfr" plot, but uses the plot method for the penalized functional regression results. See the documentation for pfr() in the refund library and for plot.gam() in the mgcv library for more information.
For a "coefs" plot, the three important functional coefficients in the model (intercept for predicting M, effect of X on M, and the effect of M on Y adjusting for X) are plotted one after another. That is, the plots are shown for the alpha_int_estimate, alpha_X_estimate, and beta_M_estimate, each as a function of time_grid. Approximate pointwise 95 percent confidence intervals are also shown if possible. If there is only one dichotomous treatment variable and panes are being used, the lower right pane will be free, so the indirect effect will be printed there even though it is a scalar.
For a "tvem" plot, the functional coefficients in the TVEM model predicting M from X are displayed.
Default is .05 for pointwise 95 percent confidence intervals.
Further arguments currently not supported