1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
|
\name{B_01_xyplot.ts}
\title{Time series plotting methods}
\alias{xyplot.ts}
\usage{
\method{xyplot}{ts}(x, data = NULL,
screens = if (superpose) 1 else colnames(x),
\dots,
superpose = FALSE,
cut = FALSE,
type = "l",
col = NULL,
lty = NULL,
lwd = NULL,
pch = NULL,
cex = NULL,
fill = NULL,
auto.key = superpose,
panel = if (superpose) "panel.superpose"
else "panel.superpose.plain",
par.settings = list(),
layout = NULL, as.table = TRUE,
xlab = "Time", ylab = NULL,
default.scales = list(y = list(relation =
if (missing(cut)) "free" else "same")))
}
\description{
This function handles time series plotting, including cut-and-stack plots.
Examples are given of superposing, juxtaposing and styling
different time series.
}
\arguments{
\item{x}{
an object of class \code{\link{ts}}, which may be multi-variate,
i.e. have a matrix structure with multiple columns.
}
\item{data}{
not used, and must be left as \code{NULL}.
}
\item{\dots}{
additional arguments passed to \code{\link{xyplot}},
which may pass them on to \code{\link{panel.xyplot}}.
}
\item{screens}{
factor (or coerced to factor) whose levels specify which
panel each series is to be plotted in. \code{screens = c(1, 2, 1)}
would plot series 1, 2 and 3 in panels 1, 2 and 1. May also be a
named list, see Details below.
}
\item{superpose}{
overlays all series in one panel (via \code{screens = 1}) and uses
grouped style settings (from
\code{trellis.par.get("superpose.line")}, etc). Note that this is
just a convenience argument: its only action is to change the
default values of other arguments.
}
\item{cut}{
defines a cut-and-stack plot. \code{cut} can be a \code{list} of
arguments to the function \code{\link{equal.count}},
i.e. \code{number} (number of intervals to divide into)
and \code{overlap} (the fraction of overlap between cuts, default
0.5). If \code{cut} is numeric this is passed as the
\code{number} argument.
\code{cut = TRUE} tries to choose an appropriate number of cuts (up
to a maximum of 6), using \code{\link{banking}}, and assuming a square
plot region. This should have the effect of minimising wasted space
when \code{aspect = "xy"}.
}
\item{type, col, lty, lwd, pch, cex, fill}{
graphical arguments, which are processed and eventually passed to
\code{\link{panel.xyplot}}.
These arguments can also be vectors or (named) lists, see Details
for more information.
}
\item{auto.key}{
a logical, or a list describing how to draw a key. See the
\code{auto.key} entry in \code{\link{xyplot}}. The default here is
to draw lines, not points, and any specified style arguments should
show up automatically.
}
\item{panel}{
the panel function. It is recommended to leave this alone, but one
can pass a \code{panel.groups} argument which is handled by
\code{\link{panel.superpose}} for each series.
}
\item{par.settings}{
style settings beyond the standard \code{col}, \code{lty},
\code{lwd}, etc; see \code{\link{trellis.par.set}} and
\code{\link{simpleTheme}}.
}
\item{layout}{
numeric vector of length 2 specifying number of columns and rows in
the plot. The default is to fill columns with up to 6 rows.
}
\item{as.table}{
to draw panels from top to bottom. The order is determined by the
order of columns in \code{x}.
}
\item{xlab, ylab}{
X axis and Y axis labels; see \code{\link{xyplot}}. Note in
particular that \code{ylab} may be a character vector, in which case the
labels are spaced out equally, to correspond to the panels;
but \emph{NOTE} in this case the vector should be reversed OR the
argument \code{as.table} set to \code{FALSE}.
}
\item{default.scales}{
\code{scales} specification. The default is set to have
\code{"free"} Y axis scales unless \code{cut} is given.
Note, users should pass the \code{scales} argument rather than
\code{default.scales}.
}
}
\details{
The handling of several graphical parameters is more
flexible for multivariate series. These parameters can be
vectors of the same length as the number of series plotted or
are recycled if shorter. They can also be (partially) named list, e.g.,
\code{list(A = c(1,2), c(3,4))} in which \code{c(3, 4)} is the
default value and \code{c(1, 2)} the value only for series \code{A}.
The \code{screens} argument can be specified in a similar way.
%% TODO: explain details of style settings, groups, etc?
Some examples are given below.
}
\value{
An object of class \code{"trellis"}. The
\code{\link[lattice:update.trellis]{update}} method can be used to
update components of the object and the
\code{\link[lattice:print.trellis]{print}} method (usually called by
default) will plot it on an appropriate plotting device.
}
\author{
Gabor Grothendieck, Achim Zeileis, Deepayan Sarkar and Felix Andrews
\email{felix@nfrac.org}.
The first two authors developed \code{xyplot.ts} in their \pkg{zoo}
package, including the \code{screens} approach. The third author
developed a different \code{xyplot.ts} for cut-and-stack plots in the
\pkg{latticeExtra} package. The final author fused these together.
}
\references{
Sarkar, Deepayan (2008) \emph{Lattice: Multivariate Data
Visualization with R}, Springer.
\url{http://lmdvr.r-forge.r-project.org/} (cut-and-stack plots)
}
\seealso{
\code{\link{xyplot}},
\code{\link{panel.xyplot}},
\code{\link{plot.ts}},
\code{\link{ts}},
\code{\link[zoo]{xyplot.zoo}} in the \pkg{zoo} package.
}
\examples{
xyplot(ts(c(1:10,10:1)))
### Figure 14.1 from Sarkar (2008)
xyplot(sunspot.year, aspect = "xy",
strip = FALSE, strip.left = TRUE,
cut = list(number = 4, overlap = 0.05))
### A multivariate example; first juxtaposed, then superposed
xyplot(EuStockMarkets, scales = list(y = "same"))
xyplot(EuStockMarkets, superpose = TRUE, aspect = "xy", lwd = 2,
type = c("l","g"), ylim = c(0, max(EuStockMarkets)))
### Examples using screens (these two are identical)
xyplot(EuStockMarkets, screens = c(rep("Continental", 3), "UK"))
xyplot(EuStockMarkets, screens = list(FTSE = "UK", "Continental"))
### Automatic group styles
xyplot(EuStockMarkets, screens = list(FTSE = "UK", "Continental"),
superpose = TRUE)
xyplot(EuStockMarkets, screens = list(FTSE = "UK", "Continental"),
superpose = TRUE, xlim = extendrange(1996:1998),
par.settings = standard.theme(color = FALSE))
### Specifying styles for series by name
xyplot(EuStockMarkets, screens = list(FTSE = "UK", "Continental"),
col = list(DAX = "red", FTSE = "blue", "black"), auto.key = TRUE)
xyplot(EuStockMarkets, screens = list(FTSE = "UK", "Continental"),
col = list(DAX = "red"), lty = list(SMI = 2), lwd = 1:2,
auto.key = TRUE)
### Example with simpler data, few data points
set.seed(1)
z <- ts(cbind(a = 1:5, b = 11:15, c = 21:25) + rnorm(5))
xyplot(z, screens = 1)
xyplot(z, screens = list(a = "primary (a)", "other (b & c)"),
type = list(a = c("p", "h"), b = c("p", "s"), "o"),
pch = list(a = 2, c = 3), auto.key = list(type = "o"))
}
\keyword{hplot}
\keyword{ ts }
|
