R: Plot diagnostics for an 'asreml' object

plot.asreml {asreml}

R Documentation

Plot diagnostics for an `asreml` object

Description

Four plots are generated: a histogram of the residuals, a Normal Q-Q plot, a plot of residuals against fitted values and a plot of residuals against unit number.

Usage

## S3 method for class 'asreml'
plot(
  x,
  res = "default",
  spatial = "trend",
  facet = TRUE,
  distribution = "norm",
  conf = 0.95,
  labels = FALSE,
  no_semi_transparency = FALSE,
  ...
)

Arguments

`x`	An `asreml` object.
`res`	The type of residuals to be plotted; see `residuals.asreml` (default = `default`).
`spatial`	If `"plot"` and an independent error has been fitted with `units` in the random formula, these are added to the residuals, otherwise if `"trend"` then `units` are not added even if present in the model (default = `"trend"`).
`facet`	If `TRUE`, multi-panel conditioning plots are produced for models with multi-section residual structures (default = `TRUE`).
`distribution`	A character string specifying the distribution for Q-Q plots. Density and quantile functions for many standard probability distributions are available in the `"stats"` package. The distribution name is automatically prefixed with `"d"` and `"q"` (default = `"norm"`).
`conf`	Value specifying confidence level (0 to 1) to be used to present confidence intervals (default = `0.95`).
`labels`	If `TRUE`, points are labelled by unit number (default = `FALSE`).
`no_semi_transparency`	This should be set to `TRUE` if the device being used does not support semi-transparency (default = `FALSE`).
`...`	Additional arguments.

Details

If the residual structure of the model contains multiple sections, the default plots are conditioned on the factor whose levels define the sections; Normal Q-Q plots are only produced if facet = TRUE. For multivariate analyses, the plots are conditioned on trait.

Value

An invisible list of ggplot2 objects.