latticeExtra (version 0.6-11)

horizonplot: Plot many time series in parallel

Description

Plot many time series in parallel by cutting the y range into segments and overplotting them with color representing the magnitude and direction of deviation.

Usage

horizonplot(x, data, ...)
## S3 method for class 'default':
horizonplot(x, data = NULL, \dots,
            panel = panel.horizonplot,
            prepanel = prepanel.horizonplot,
            strip = FALSE, strip.left = TRUE,
            par.strip.text = list(cex = 0.6),
            groups = NULL,
            default.scales =
              list(y = list(relation = "sliced", axs = "i", 
                            draw = FALSE, tick.number = 2)))
panel.horizonplot(x, y, ..., origin,
            border = NA, col.regions = regions$col)
prepanel.horizonplot(x, y, ..., origin = function(y) na.omit(y)[1])

Arguments

x, y

Argument on which argument dispatch is carried out. Typically this will be a multivariate time series.

In the panel and prepanel functions, these are the data coordinates.

data

Not used (at least, not used by xyplot.ts).

...

further arguments. Arguments to xyplot as well as to the default panel function panel.horizonplot can be supplied directly to horizonplot.

In typical usage,

panel

function to render the graphic given the data. This is the function that actually implements the display.

prepanel

function determining range of the data rectangle from data to be used in a panel.

strip, strip.left

by default strips are only drawn on the left, to save space.

par.strip.text

graphical parameters for the strip text; see xyplot. One notable argument here is lines, allowing multi-line text.

default.scales

see xyplot.

groups

not applicable to this type of plot.

origin

the baseline y value for the first (positive) segment. This can be a number, which is then fixed across all panels, or it can be a function, which is evaluated with the y values in each panel. The default is the first non-missing

border

border color for the filled polygons, defaults to no border.

col.regions

color scale, taken by default from the current theme, trellis.par.get("regions")$col. This should be a divergent color scale (with white as the central color).

Value

• An object of class "trellis". The update method can be used to update components of the object and the print method (usually called by default) will plot it on an appropriate plotting device.

Details

This function draws time series as filled areas, with modifications to effectively visualise many time series in parallel. Data that would be drawn off the top of each panel is redrawn from the bottom of the panel in a darker color. Values below the origin are inverted and drawn in the opposite color. There are up to three shades (typically in blue) for data above the baseline and up to three shades (typically in red) for data below the baseline. See the article referenced below for an introduction to Horizon plots.

Note that the y scale in each panel defines the actual origin and scale used. The origin argument is only used in the prepanel function to choose an appropriate y scale (it is ignored by the panel function). The ylim argument therefore over-rides origin. The choice of scales$y$relation is also crucial; see xyplot and the examples below.

References

Stephen Few (2008). Time on the Horizon. Visual Business Intelligence Newsletter, June/July 2008 http://www.perceptualedge.com/articles/visual_business_intelligence/time_on_the_horizon.pdf

See Also

Lattice, xyplot.ts, panel.xyarea

Examples

## generate a random time series object with 12 columns
set.seed(1)
dat <- ts(matrix(cumsum(rnorm(200 * 12)), ncol = 12))
colnames(dat) <- paste("series", LETTERS[1:12])

## show simple line plot first, for reference.
xyplot(dat, scales = list(y = "same"))

## horizonplot:
## draw each series with a different baseline;
## ('origin' default is the first data value in each series).
## This highlights deviations within each series.
horizonplot(dat, layout = c(1,12))

## now use same scales in each panel
## (so colors represent common ranges across all panels).
## This highlights differences across series.
## One could potentially draw a colorkey in this case.
horizonplot(dat, scales = list(y = list(relation = "same")),
            origin = 0, layout = c(1,12)) +
  layer(panel.scaleArrow(x = .99, pos = 2))

## illustration of the cut points used in the following plot
xyplot(EuStockMarkets, scales = list(y = "same"),
  panel = function(x, y, ...) {
    rgn <- trellis.par.get("regions")$col
    col <- rgn[1 + (length(rgn)-1) * 0:7/7]
    for (i in c(-4:-1, 3:0)) {
      if (i >= 0)
        yi <- pmax(4000, pmin(y, 4000 + 1000 * (i+1)))
      if (i < 0)
        yi <- pmin(4000, pmax(y, 4000 + 1000 * i))
      panel.xyarea(x, yi, origin = 4000,
        col = col[i+5], border = NA)
    }
    panel.lines(x, y)
    panel.abline(h = 4000, lty = 2)
  })

## compare with previous plot
horizonplot(EuStockMarkets, strip.left = TRUE,
  scales = list(y = list(relation = "same")), ylim = c(4000, 5000)) +
layer(panel.scaleArrow(x = 0.99, col = "grey", srt = 90, cex = 0.75))

.featured_example <- TRUE
## a cut-and-stack plot; should use "same" y scales!
horizonplot(sunspots, cut = list(n = 23, overlap = 0),
  scales = list(y = list(relation = "same"), draw = FALSE),
  origin = 100, strip.left = FALSE, layout = c(1,23)) +
layer(grid::grid.text(round(x[1]), x = 0, just = "left"))