CDF bar plot for model predictions

mc_obs_cdf(...)

mc_model_cdf(..., n_sample = NA, group_sample = "collapse")

Arguments

...: Augments passed to ggdist::stat_cdfinterval.
n_sample: The number of sample of draws to show in CDF bar plot. Default NA. If n_sample is NA, then mc_model_cdf() will use all draws from posterior distribution.
group_sample: How to show samples? Default "collapse". If group_sample is "collapse", then all samples are collapsed together to show in one CDF bar plot; if group_sample is "group", then each sample is shown in an individual CDF bar plot; if group_sample is "hops", then mc_model_cdf() will use animation to show each sample in one frame; if group_sample is an function, then all samples are aggregated by group_sample().

Examples

library(ggplot2)

mcplot(mpg_model) +
  mc_model_cdf(n_sample = 50) +
  mc_obs_cdf() +
  mc_gglayer(coord_flip())


mcplot(mpg_model) +
  mc_model_cdf(n_sample = 50) +
  mc_obs_cdf() +
  mc_condition_on(x = vars(vs))


mcplot(mpg_model) +
  mc_model_cdf(n_sample = 50, group_sample = "group") +
  mc_obs_cdf() +
  mc_condition_on(x = vars(vs))


mcplot(mpg_model) +
  mc_model_cdf(n_sample = 50, group_sample = mean) +
  mc_observation_transformation(mean) +
  mc_obs_cdf() +
  mc_condition_on(x = vars(vs))