Learn R Programming

clarify (version 0.2.1)

plot.clarify_setx: Plot marginal predictions from sim_setx()

Description

plot.clarify_sext() plots the output of sim_setx(), providing graphics similar to those of plot.clarify_est() but with features specifically for plot marginal predictions. For continues predictors, this is a plot of the marginal predictions and their confidence bands across levels of the predictor. Otherwise, this is is a plot of simulated sampling distribution of the marginal predictions.

Usage

# S3 method for clarify_setx
plot(
  x,
  var = NULL,
  ci = TRUE,
  level = 0.95,
  method = "quantile",
  reference = FALSE,
  ...
)

Value

A ggplot object.

Arguments

x

a clarify_est object resulting from a call to sim_setx().

var

the name of the focal varying predictor, i.e., the variable to be on the x-axis of the plot. All other variables with varying set values will be used to color the resulting plot. See Details. Ignored if no predictors vary or if only one predictor varies in the reference grid or if x1 was specified in sim_setx(). If not set, will use the predictor with the greatest number of unique values specified in the reference grid.

ci

logical; whether to display confidence intervals or bands for the estimates. Default is TRUE.

level

the confidence level desired. Default is .95 for 95% confidence intervals.

method

the method used to compute confidence intervals or bands. Can be "wald" to use a Normal approximation or "quantile" to use the simulated sampling distribution (default). See summary.clarify_est() for details. Abbreviations allowed.

reference

logical; whether to overlay a normal density reference distribution over the plots. Default is FALSE. Ignored when variables other than the focal varying predictor vary.

...

for plot(), further arguments passed to ggplot2::geom_density().

Details

plot() creates one of two kinds of plots depending on how the reference grid was specified in the call to sim_setx() and what var is set to. When the focal varying predictor (i.e., the one set in var) is numeric and takes on three or more unique values in the reference grid, the produced plot is a line graph displaying the value of the marginal prediction (denoted as E[Y|X]) across values of the focal varying predictor, with confidence bands displayed when ci = TRUE. If other predictors also vary, lines for different values will be displayed in different colors. These plots are produced using ggplot2::geom_line() and ggplot2::geom_ribbon()

When the focal varying predictor is a factor or character or only takes on two or fewer values in the reference grid, the produced plot is a density plot of the simulated predictions, similar to the plot resulting from plot.clarify_est(). When other variables vary, densities for different values will be displayed in different colors. These plots are produced using ggplot2::geom_density().

Marginal predictions are identified by the corresponding levels of the predictors that vary. The user should keep track of whether the non-varying predictors are set at specified or automatically set "typical" levels.

See Also

summary.clarify_est() for computing p-values and confidence intervals for the estimated quantities.

Examples

Run this code
## See help("sim_setx") for examples

Run the code above in your browser using DataLab