Re: lme: Plot colours based on variable?

On Tue, 2 Dec 2025 at 21:56, Ben Bolker <[email protected]> wrote:

>    I often find custom-built lattice-based plot methods inscrutable and
> end up falling back to figuring out how to get what I want directly with
> lattice.
>

The documentation suggests that this might work, but the colors come out
weird.

plot(BodyWeight, innerGroups = ~Diet)


>
>    I haven't figured out how to get the label 'Diet' on the legend, but
> otherwise this works pretty well ...
>
>
> library(lattice)
> library(nlme)
> labs <- paste(attr(BodyWeight, "labels"),
>                attr(BodyWeight, "units"))
> xyplot(weight ~ Time | Rat, type = c("p", "r"),
>         data = BodyWeight,
>         layout = c(NA, 1),  ## enforce 1 row
>

or aspect = "xy", as in the plot method.


>         groups = Diet,
>         auto.key = TRUE,
>         pch = 19,
>         xlab = labs[1],
>         ylab = labs[2],
>         par.settings = list(superpose.line = list(col = "black")))
>

A more direct analog of plot(BodyWeight) would be

xyplot(BodyWeight, weight ~ Time | Rat,
       aspect = "xy", type = "o",
       groups = Diet, auto.key = list(title = "Diet", columns = 3),
       grid = TRUE, xlab = labs[1], ylab = labs[2])

-Deepayan



>
>
> On 12/2/25 06:06, Duncan Murdoch wrote:
> > On 2025-12-02 3:39 a.m., Leo Mada via R-help wrote:
> >> Dear R-Users,
> >>
> >> Is there a quick solution to plot colours based on a variable?
> >>
> >> I tried something like this:
> >>
> >> library(nlme)
> >>
> >> plot(BodyWeight)
> >>
> >> # Does NOT work!
> >> colDiet = c(2:4)[BodyWeight$Diet]
> >> plot(BodyWeight, col = colDiet)
> >>
> >> # Wastes space
> >> rat.fit = lme(weight ~ Time + Diet, data = BodyWeight, random = ~
> >> Time|Rat)
> >> plot(rat.fit, weight ~ Time | Rat + Diet)
> >>
> >> Is there a quick way to base the colour on the Diet variable?
> >
> >    plot(BodyWeight, ...)
> >
> > dispatches to plot.nfnGroupedData(), which uses lattice::xyplot()
> > graphics rather than base graphics. I don't know lattice well enough to
> > answer your question, but maybe this will give you a hint...
> >
> > Duncan Murdoch
> >
> > ______________________________________________
> > [email protected] mailing list -- To UNSUBSCRIBE and more, see
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide https://www.R-project.org/posting-
> > guide.html
> > and provide commented, minimal, self-contained, reproducible code.
>
> --
> Dr. Benjamin Bolker
> Professor, Mathematics & Statistics and Biology, McMaster University
> Associate chair (graduate), Mathematics & Statistics
> Director, School of Computational Science and Engineering
> * E-mail is sent at my convenience; I don't expect replies outside of
> working hours.
>
> ______________________________________________
> [email protected] mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> https://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

	[[alternative HTML version deleted]]