plot.CEAforests: Plotting function for CEA forests.

In bonander/CEAforests: Tree-based heterogenous cost-effectiveness and policy learning with individual-level data (beta)

Description

Provides histograms, scatter plots or (partial) effects plots to assess heterogeneity with respect to a covariate after a CEAforest.

Usage

## S3 method for class 'CEAforests'
plot(
  forest,
  which.y = "all",
  which.x = NULL,
  conditional = FALSE,
  smooth = FALSE,
  ci.level = 0.95,
  x.range = NULL,
  length.out = 100,
  xlab = "X",
  labels = "AUTO",
  ...
)

Arguments

forest	A trained CEA forest.
which.y	A string or string vector naming which outcomes to plot (any combination of "outcome", "cost" and "nmb" is acceptable). Defaults to "all", which produces three plots in one-row, three-column grid.
which.x	A column number or string naming a single variable from the X matrix in the CEAforests object. If null, the function outputs histograms of the out-of-bag estimates.
conditional	Whether or not to keep all other variables in the X matrix constant at their mean in bivariate plots (defaults to FALSE). Ignored if which.x is null.
smooth	Whether or not to plot a semi-parametric smooth function fit to doubly robust scores for tau(x) (via the mgcv package) instead of out-of-bag estimates (unconditional) or non-parametric predictions (conditional). Defaults to FALSE.
ci.level	The desired confidence level for confidence intervals (used when applicable). Defaults to 0.95.
x.range	A two-element numeric vector that controls the range of the x-axis. Defaults to min(which.x) and max(which.x).
length.out	The length of the sequence of X values to be plotted (used when conditional=TRUE). Defaults to 100.
xlab	Label for the X axis. Defaults to the column name for the focal X variable, or to "X" if no column name is available.
labels	Labels to be passed to cowplot::plot_grid when more than one outcome type (which.y) is supplied. Defaults to "AUTO", which labels the plots using letters.
...	Additional arguments to be passed to the gam function in mgcv. Ignored if smooth=FALSE.

Value

A ggplot via ggplot2 or grid of ggplots via the cowplot package.

Examples

## Not run: 
To be added...

## End(Not run)

bonander/CEAforests documentation built on April 1, 2021, 10:57 a.m.