plot.summary_fitsae function

Plot Method for a summary_fitsae Object

The generic method plot() provides, in a grid (default) or sequence, (a) a scatterplot of direct estimates versus model-based estimates, visually capturing the shrinking process, (b) a Bayesian P-values histogram, (c) a boxplot of standard deviation reduction values, and, if areas sample sizes are provided as input in fit_sae(), (d) a scatterplot of model residuals versus sample sizes, in order to check for design-consistency i.e., as long as sizes increase residuals should converge to zero.

## S3 method for class 'summary_fitsae'
plot(
  x,
  size = 2.5,
  alpha = 0.8,
  n_bins = 15,
  grid = TRUE,
  label_names = NULL,
  ...
)

Arguments

• x: Object of class summary_fitsae.
• size: Aesthetic option denoting the size of scatterplots points, see geom_point documentation.
• alpha: Aesthetic option denoting the opacity of scatterplots points, see geom_point documentation.
• n_bins: Denoting the number of bins used for histogram.
• grid: Logical indicating whether plots are displayed in a grid (TRUE) or in sequence (FALSE).
• label_names: Character string indicating the model name to display in boxplot x-axis label.
• ...: Currently unused.

Returns

Four ggplot2 objects in a grid.

Examples

library(tipsae)

# loading toy dataset
data("emilia_cs")

# fitting a model
fit_beta <- fit_sae(formula_fixed = hcr ~ x, data = emilia_cs, domains = "id",
                    type_disp = "var", disp_direct = "vars", domain_size = "n",
                    # MCMC setting to obtain a fast example. Remove next line for reliable results.
                    chains = 1, iter = 150, seed = 0)

# check model diagnostics
summ_beta <- summary(fit_beta)

# visualize diagnostics via plot() method
plot(summ_beta)

See Also

summary.fitsae to produce the input object.