R: Plot an nfnGroupedData Object

plot.nfnGroupedData {nlme}

R Documentation

Plot an nfnGroupedData Object

Description

A Trellis plot of the response versus the primary covariate is generated. If outer variables are specified, the combination of their levels are used to determine the panels of the Trellis display. Otherwise, the levels of the grouping variable determine the panels. A scatter plot of the response versus the primary covariate is displayed in each panel, with observations corresponding to same inner group joined by line segments. The Trellis function xyplot is used.

Usage

## S3 method for class 'nfnGroupedData'
plot(x, outer, inner, innerGroups, xlab, ylab, strip, aspect, panel,
     key, grid, ...)

Arguments

`x`	an object inheriting from class `nfnGroupedData`, representing a `groupedData` object with a numeric primary covariate and a single grouping level.
`outer`	an optional logical value or one-sided formula, indicating covariates that are outer to the grouping factor, which are used to determine the panels of the Trellis plot. If equal to `TRUE`, `attr(object, "outer")` is used to indicate the outer covariates. An outer covariate is invariant within the sets of rows defined by the grouping factor. Ordering of the groups is done in such a way as to preserve adjacency of groups with the same value of the outer variables. Defaults to `NULL`, meaning that no outer covariates are to be used.
`inner`	an optional logical value or one-sided formula, indicating a covariate that is inner to the grouping factor, which is used to associate points within each panel of the Trellis plot. If equal to `TRUE`, `attr(object, "inner")` is used to indicate the inner covariate. An inner covariate can change within the sets of rows defined by the grouping factor. Defaults to `NULL`, meaning that no inner covariate is present.
`innerGroups`	an optional one-sided formula specifying a factor to be used for grouping the levels of the `inner` covariate. Different colors, or line types, are used for each level of the `innerGroups` factor. Default is `NULL`, meaning that no `innerGroups` covariate is present.
`xlab`, `ylab`	optional character strings with the labels for the plot. Default is the corresponding elements of `attr(object, "labels")` and `attr(object, "units")` pasted together.
`strip`	an optional function passed as the `strip` argument to the `xyplot` function. Default is `strip.default(..., style = 1)` (see `trellis.args`).
`aspect`	an optional character string indicating the aspect ratio for the plot passed as the `aspect` argument to the `xyplot` function. Default is `"xy"` (see `trellis.args`).
`panel`	an optional function used to generate the individual panels in the Trellis display, passed as the `panel` argument to the `xyplot` function.
`key`	an optional logical function or function. If `TRUE` and `innerGroups` is non-`NULL`, a legend for the different `innerGroups` levels is included at the top of the plot. If given as a function, it is passed as the `key` argument to the `xyplot` function. Default is `TRUE` if `innerGroups` is non-`NULL` and `FALSE` otherwise.
`grid`	an optional logical value indicating whether a grid should be added to plot. Default is `TRUE`.
`...`	optional arguments passed to the `xyplot` function.

Value

a Trellis plot of the response versus the primary covariate.

Author(s)

José Pinheiro and Douglas Bates bates@stat.wisc.edu

References

Bates, D.M. and Pinheiro, J.C. (1997), "Software Design for Longitudinal Data", in "Modelling Longitudinal and Spatially Correlated Data: Methods, Applications and Future Directions", T.G. Gregoire (ed.), Springer-Verlag, New York.

Examples

# different panels per Subject
plot(Orthodont)
# different panels per gender
plot(Orthodont, outer = TRUE)

[Package nlme version 3.1-166 Index]