lmer model - accessing elements of the model

Question

I am an R beginner and even newer to lmer and mixed effects modelling.

If I have an lmer model of:

model <- lmer(C ~ Group * Time + (1|ID), REML = FALSE, data=data)

How do I access details of the model in order to plot them?

I know I can plot the residuals against the fitted values using either:

plot(model)

OR

plot(predict(model), resid(model))

But is there a way to access other elements/details of the model? What other model elements can be accessed? I am aware that I can use the function 'summary()' to look up the random and fixed effect. But what about plotting the residuals against other elements of the model e.g. residuals against fixed effects or random effect?

Answer 1

When you have doubts like this you shoud check the Vignette of the package, in this case vignette(lme4). Only with a superficial search I found this table that maybe is of use:

Method	Description
anova	Decomposition of fixed-effects contributions or model comparison.
as.function	Function returning profiled deviance or REML criterion.
coef	Sum of the random and fixed effects for each level.
confint	Confidence intervals on linear mixed-model parameters.
deviance	Minus twice maximum log-likelihood. (Use REMLcrit for the REML criterion.)
df.residual	Residual degrees of freedom.
drop1	Drop allowable single terms from the model.
extractAIC	Generalized Akaike information criterion
fitted	Fitted values given conditional modes.
fixef	Estimates of the fixed-effects coefficients
formula	Mixed-model formula of fitted model.
logLik	Maximum log-likelihood.
model.frame	Data required to fit the model.
model.matrix	Fixed-effects model matrix
ngrps	Number of levels in each grouping factor.
nobs	Number of observations.
plot	Diagnostic plots for mixed-model fits.
predict	Various types of predicted values.
print	Basic printout of mixed-model objects.
profile	Profiled likelihood over various model parameters.
ranef	Conditional modes of the random effects.
refit	A model (re)fitted to a new set of observations of the response variable.
refitML	A model (re)fitted by maximum likelihood.
residuals	Various types of residual values.
sigma	Residual standard deviation.
simulate	Simulated data from a fitted mixed model.
summary	Summary of a mixed model.
terms	Terms representation of a mixed model.
update	An updated model using a revised formula or other arguments.
VarCorr	Estimated random-effects variances, standard deviations, and correlations.
vcov	Covariance matrix of the fixed-effect estimates.
weights	Prior weights used in model fitting.

Answer 2

Here's one way to plot residuals vs your time factor:

library(broom.mixed)
aa <- augment(model)
boxplot(.resid ~ Time, data = aa)

You could also use the built-in plot method (see help("plot.merMod")):

plot(model, resid(.) ~ Time)

Answer 3

Yes you can use summary() function just like with any other lm or glm object to get an assessment of the actual model.

Similarly you can treat these objects like you would a dataframe.

The only difference is instead of using the $ you should use the @ symbol.

For example:

lm <- lmer(Petal.Length ~ Sepal.Length * Sepal.Width + (1|Species), data = iris)

lm@frame %>% head()

 Petal.Length Sepal.Length Sepal.Width Species
1          1.4          5.1         3.5  setosa
2          1.4          4.9         3.0  setosa
3          1.3          4.7         3.2  setosa
4          1.5          4.6         3.1  setosa
5          1.4          5.0         3.6  setosa
6          1.7          5.4         3.9  setosa

You can from that point explore deeper in the object.