residplot: Diagnostic Plots for a Linear (Mixed) Model

In predictmeans: Predicted Means for Linear and Semiparametric Models

Diagnostic Plots for a Linear (Mixed) Model

Description

This function produces diagnostic plots for linear models including 'aov', 'lm', 'glm', 'gls', 'lme' and 'lmer'.

Usage

  residplot(model, group = "none", level = 1, slope = FALSE, id = FALSE, newwd=TRUE,
    ask=FALSE)

Arguments

model	Model object returned by aov, lm, glm, gls, lme, and lmer.
group	Name (in "quotes") for indicating the variable used to show grouping in the residual vs predicted plot. If variable is a term in the model, then group will be a name of the variable such as group="A", otherwise group will be the actual variable such as group=data$A.
level	An integer 1, 2, etc. used to specify a level of the random effect for plotting. The default value is 1.
slope	A logical variable. If set to TRUE, a Q-Q plot of random slope will be drawn.
id	A logical variable. If set to TRUE, outliers in the residual vs fitted plot can be identified interactively.
newwd	A logical variable to indicate whether to print graph in a new window. The default is TRUE.
ask	logical. If TRUE (and the R session is interactive) the user is asked for input, before a new figure is drawn.

Author(s)

Dongwen Luo, Siva Ganesh and John Koolaard

Examples

## Note that the order of levels of nested random effects is oposite 
## between lme and lmer objects.

library(predictmeans)
Oats$nitro <- factor(Oats$nitro)
fm <- lme(yield ~ nitro*Variety, random=~1|Block/Variety, data=Oats)
residplot(fm, level=2)    #lme: level=2 for random effect "Block:Variety"

#  Not Run
#  library(lme4)
#  fm <- lmer(yield ~ nitro*Variety+(1|Block/Variety), data=Oats)
#  residplot(fm) # lmer: By default level=1 for random effect "Block:Variety"

predictmeans documentation built on Oct. 20, 2023, 5:07 p.m.