plot.galamm: Diagnostic plots for galamm objects

In galamm: Generalized Additive Latent and Mixed Models

Diagnostic plots for galamm objects

Description

This function provides diagnostic plots for models fitted with galamm(). See the residuals.galamm() function for definition of the residuals being used.

Usage

## S3 method for class 'galamm'
plot(x, form = resid(., type = "pearson") ~ fitted(.), abline = NULL, ...)

Arguments

x	An object of class galamm returned from galamm.
form	An option formula specifying the desired type of plot. Conditioning variables are specified with a vertical bar.
abline	An optional numeric vector specifying the intercept and slope of a line to be added to the plot.
...	Optional arguments passed on to the plot functions in the lattice package.

Details

The interface of this function is designed to be similar to the plot.merMod function from lme4 \insertCitebatesFittingLinearMixedEffects2015galamm.

Value

A plot is displayed.

Author(s)

Douglas Bates, Martin Maechler, Ben Bolker, and Steven Walker, with modifications by Øystein Sørensen.

References

\insertAllCited

See Also

residuals.galamm() for extracting residuals and plot() for the generic function.

Other diagnostics: qqmath.galamm()

Examples

## Linear mixed model example from lme4
data("sleepstudy", package = "lme4")
mod <- galamm(Reaction ~ Days + (Days | Subject), data = sleepstudy)

# Diagnostic plot of Pearson residuals versus fitted values
plot(mod)

# Include a straight line at zero
plot(mod, abline = c(0, 0))

# Diagnostic plot of Pearson residuals versus fitted values per subject
plot(mod, form = resid(., type = "pearson") ~ fitted(.) | Subject)

# Residuals plotted against time with a straight line at zero
plot(mod, form = resid(., type = "pearson") ~ Days, abline = c(0, 0))

# Residuals plotted against time per subject with a straight line at zero
plot(mod, form = resid(., type = "pearson") ~ Days | Subject,
     abline = c(0, 0))

# Box plot of residuals by Subject
plot(mod, Subject ~ resid(., scaled=TRUE))

## Logistic mixed model example from lme4
data("cbpp", package = "lme4")
mod <- galamm(cbind(incidence, size - incidence) ~ period + (1 | herd),
              data = cbpp, family = binomial)

# Diagnostic plot using Pearson residuals
plot(mod)

# Diagnostic plot using deviance residuals
plot(mod, resid(., type = "deviance") ~ fitted(.))

# Diagnostic plot per herd
plot(mod, resid(., type = "deviance") ~ fitted(.) | herd)

## Linear mixed model with factor structures
# See vignette on linear mixed models with factor structures for details
data(KYPSsim, package = "PLmixed")
KYPSsim <- KYPSsim[KYPSsim$sid < 100, ]
KYPSsim$time <- factor(KYPSsim$time)

loading_matrix <- rbind(c(1, 0), c(NA, 0), c(NA, 1), c(NA, NA))
factors <- c("ms", "hs")
load_var <- "time"
form <- esteem ~ time + (0 + ms | mid) + (0 + hs | hid) + (1 | sid)

mod <- galamm(formula = form, data = KYPSsim, factor = factors,
              load_var = load_var, lambda = loading_matrix)

# Pearson residuals plotted against fitted value
plot(mod)

# Actual value plotted against fitted value with a line crossing the diagonal
plot(mod, form = esteem ~ fitted(.), abline = c(0, 1))

galamm documentation built on Dec. 21, 2025, 5:07 p.m.