spec_plot: Plot Posterior Quantities of Interest Under Alternative Model...

In CoryMcCartan/adjustr: Stan Model Adjustments and Sensitivity Analyses using Importance Sampling

Plot Posterior Quantities of Interest Under Alternative Model Specifications

Description

Uses weights computed in adjust_weights to plot posterior quantities of interest versus specification parameters

Usage

spec_plot(
  x,
  by,
  post,
  only_mean = FALSE,
  ci_level = 0.8,
  outer_level = 0.95,
  ...
)

Arguments

x	An adjustr_weighted object.
by	The x-axis variable, which is usually one of the specification parameters. Can be set to 1 if there is only one specification. Automatically quoted and evaluated in the context of x.
post	The posterior quantity of interest, to be computed for each resampled draw of each specification. Should evaluate to a single number for each draw. Automatically quoted and evaluated in the context of x.
only_mean	Whether to only plot the posterior mean. May be more stable.
ci_level	The inner credible interval to plot. Central 100*ci_level posterior draws.
outer_level	The outer credible interval to plot.
...	Ignored.

Value

A ggplot object which can be further customized with the ggplot2 package.

See Also

adjust_weights, summarize.adjustr_weighted

Examples

## Not run: 
spec = make_spec(eta ~ student_t(df, 0, scale),
                 df=1:10, scale=seq(2, 1, -1/9))
adjusted = adjust_weights(spec, eightschools_m)

spec_plot(adjusted, df, theta[1])
spec_plot(adjusted, df, mu, only_mean=TRUE)
spec_plot(adjusted, scale, tau)

## End(Not run)

CoryMcCartan/adjustr documentation built on June 25, 2022, 4:44 p.m.