plot.mcglm: Diagnostic Plots for mcglm Objects

In mcglm: Multivariate Covariance Generalized Linear Models

Diagnostic Plots for mcglm Objects

Description

Produces diagnostic plots for fitted mcglm objects and returns the data used to generate the plots. Available diagnostics include:

• "residuals": plots of fitted values vs. Pearson residuals and Q-Q plots of residuals;

• "algorithm": plots of regression and covariance parameters and quasi-score iterations to inspect algorithm convergence;

• "partial_residuals": partial residual plots for each covariate in the model.

Usage

## S3 method for class 'mcglm'
plot(
  x,
  type = c("residuals", "algorithm", "partial_residuals"),
  plot_graphics = TRUE,
  ...
)

Arguments

x	A fitted object of class mcglm.
type	Character string specifying the type of diagnostic plot to produce. One of "residuals", "algorithm", or "partial_residuals".
plot_graphics	Logical; if TRUE (default), the function produces the plots. If FALSE, the function only returns the data used for plotting, which can be used to produce custom plots.
...	Additional arguments; currently ignored. Included for compatibility with the generic plot function.

Value

A list containing the data used to generate the plots:

• For "residuals": fitted values and residuals.

• For "algorithm": IterationRegression, IterationCovariance, ScoreRegression, ScoreCovariance.

• For "partial_residuals": residuals, list_beta, and partial_residuals containing x and y values for each covariate.

Author(s)

Wagner Hugo Bonat, [e-mail]

See Also

residuals, fitted, plot

Examples

library(mcglm)
set.seed(123)
mydata <- data.frame(y = rnorm(10), x1 = rnorm(10),
                    x2 = rbinom(10, size = 1, prob = 0.5))
Z0 <- mc_id(mydata)
fit <- mcglm(c(y ~ x1 + x2), matrix_pred = list(Z0), data = mydata)
# Produce plots and get data
diag_data <- plot(fit, type = "residuals")
# Only get data without plotting
diag_data <- plot(fit, type = "partial_residuals", plot_graphics = FALSE)

mcglm documentation built on Jan. 9, 2026, 1:07 a.m.