| linechart {mining} | R Documentation |
Plot each data row as a curve.
linechart(y,se=NULL,effects=F,med=F,
xscale=c("equal","linear","none"),...)
y |
a named vector or matrix. |
se |
a vector or matrix, same size as y, of error bounds.
Alternatively, y can be list(y,se). |
effects |
If TRUE, the columns are shifted to have zero
mean. |
med |
If TRUE and effects=TRUE, the columns are
shifted to have zero median. |
xscale |
describes how the columns should be placed. |
... |
additional arguments to labeled.curves. |
If xscale="linear", the columns are placed to make each curve
as straight as possible.
If xscale="equal", the columns are placed similar to
"linear" but with the constraint that they must be equally
spaced.
If xscale="none", the columns are placed in the order that
they appear in the matrix. This is automatic if y has ordered
columns (see dim.ordered).
If se != NULL, error bars are drawn around each point.
Linecharts are a replacement for dotcharts and mosaics.
Tom Minka
# compare to a dotchart
data(VADeaths)
dotchart(VADeaths, main = "Death Rates in Virginia - 1940")
dimOrdered(VADeaths)[2] = F
linechart(VADeaths)
linechart(t(VADeaths))
# compare to a mosaicplot
data(HairEyeColor)
x <- margin.table(HairEyeColor,c(1,2))
dimOrdered(x) = F
mosaicplot(x)
x = t(x)
col = c("brown","blue","red","green")
linechart(row.probs(x),color.pal=col)
linechart(row.probs(x,se=T),color.pal=col)
linechart(row.probs(x,se=T),jitter=0.02,color.pal=col)
mosaicplot(x)
linechart(row.probs(t(x),se=T))
data(blood)
dimOrdered(blood) = F
linechart(row.probs(blood,se=T))
data(antacids)
dimOrdered(antacids) = F
linechart(row.probs(antacids,se=T))
mosaicplot(t(antacids))