[R] change fitted line colour in lme() trellis plot?

[Deepayan Sarkar]

On 1/19/06, Bill Simpson <william.simpson at drdc-rddc.gc.ca> wrote:
> If I used a groupedData object, if I do
> fit<-lme(blah)
> then
> plot(augPred(fit))
> produces a nice trellis plot of the data along with the fitted lines
>
> However I find that the lines and the data points are in the same colour
> (light blue against a medium grey background). Is there a way to make
> the lines in a different colour (e.g. black)?

plot(augPred(fit), col.line = 'black')

> It would also be nice if
> the line were plotted after the points so it is visible (I have a lot of
> points and the line is obscured).

The order is hard-coded in the panel function:

> plot(augPred(fit))$panel
function (x, y, subscripts, groups, ...)
{
    if (grid)
        panel.grid()
    orig <- groups[subscripts] == "original"
    panel.xyplot(x[orig], y[orig], ...)
    panel.xyplot(x[!orig], y[!orig], ..., type = "l")
}

You can change it by supplying your own panel function, e.g.

plot(augPred(fit), col.line = "black",
     panel =
     function (x, y, subscripts, groups, ...) {
         panel.grid()
         orig <- groups[subscripts] == "original"
         panel.xyplot(x[orig], y[orig], ...)
         panel.xyplot(x[!orig], y[!orig], ..., type = "l")
     })

Deepayan