Skip to contents

Posterior predictive ECDF plot

Source: R/posterior_predict.R
plot_ppc_ecdf.Rd

Plot empirical CDFs of observed vs replicated data

Usage

plot_ppc_ecdf(
  fit,
  ppd = NULL,
  ndraws_plot = 20,
  alpha = 0.3,
  rel = 1,
  time = NULL,
  Y_obs = NULL
)

Arguments

fit

A dbn model fit object

ppd

Posterior predictive samples (from posterior_predict_dbn)

ndraws_plot

Number of replicates to show (default: 20)

alpha

Transparency for replicate lines (default: 0.3)

rel

Relation index to plot (default: 1)

time

Time index to plot (default: all)

Y_obs

Observed data array (required)

Value

A ggplot object

See also

posterior_predict_dbn, plot_ppc_density, param_summary

Examples

# \donttest{
sim <- simulate_static_dbn(n = 6, time = 5, seed = 1)
fit <- dbn(sim$Y, model = "static", nscan = 200, burn = 100, verbose = FALSE)
ppd <- posterior_predict_dbn(fit, ndraws = 5)
if (requireNamespace("ggplot2", quietly = TRUE)) plot_ppc_ecdf(fit, ppd, Y_obs = sim$Y)

# }